Spot Play Contest $500

[km]

Spot Play Contest!

$500 prize to the winner

Contestants will submit the specs for their spot play on this forum. I will run a 365-day test on it. MikeDee can run it in Access to back me up. The highest ROI wins. But, you must meet the following criteria =

= 600 more sample plays in a 365-day period

= play cannot be track specific, we must be able to run it on ALL tracks, but you can specify Purse parameters, EPR range, Vi, Class type, distance, surface, etc.

= win rate > 10%

= highest ROI that meets the criteria above gets the cash, but it must be at least ROI 1.10 (+10%) or greater return. If there is a tie, the prize money goes to the person with the greater sample size. If there is a duplicate play submitted, the prize money goes to the one who posted it first.

= contest ends Friday March 2. Winner announced and play listed in the March newsletter.

Think your spot play can meet all these requirements?
might be 5 Benjamins coming your way if you have the highest ROI

Remember this is a public posting of your play to get the cash

if no one posts a play that meets all the criteria, no prize awarded.

this is an contest to stimulate ideas, please post the spot play in text format so everyone can understand the parameters

contest open to any registered bbs user using HTR Export or Robot on HDW data files

the test verification run will be static: Feb1, 2006 - January 31, 2007
so everyone will get their play tested under identical data samples

good luck, submit as many plays as you want

[km]

Example submission

nTRN > 400
nJKY > 300
nMLO >= 8/1
nPURSE > = 10
nDist > 6
rK < 4

[Mark]

This may violate the spirit of the contest (Artificial surface only) and I am prepared for Inquiries and Objections, nonetheless at the very least I can provide a target.

Clm
3yr 3up 4up Fems Male
Purse<=>$10k Vi 10-32 No Chaos PL-5

Dist: ALL
Surf: Artif

Krnk: K1-K7

TRKS: HOL KEE TP WO

LAY: 010-015 EPR:090-115 TRN:200-550


I get appoximately 650 plays @ 1.11 WROI 16% win
The ROI improves dramatically at 5/1+ though plays drop to around 350.

[km]

Thanks for submitting the play Mark, you stayed within the rules, we'll test it with ALL tracks, but "A" surface only, so it still qualifies as long as it gets 600+ plays, I'll run a query and post the results here.

[km]

Mark's play produced a 1.13 (+13%) profit with 640 plays. Well done Mark! So far you have the $500, let's see if anyone can top you.

Play All 00640 00107 17% 31% 42% 1.13 0.98 0.89 $13.5 0020 $72 1.59

Races Tested: 611 01Feb06 - 31Jan07

Type: Clm
AgSx: 3yr 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 NO CHAOS PL-5

Dist: ALL
Surf: Artif

Krnk: K1 K2 K3 K4 K5 K6 K7

TRKS: ALL

Rank Filters:
Item Filters:

LAY: 010-015 EPR:090-115 TRN:200-500

[njcurveball]

WOW! Nice job! Who says you cannot make money on Polytrack!

[Pizzaman]

nWRK>=83
rHTR=1,2
nSURF=2 (Turf)
nSEX=1 (Males)
rML >=3

[km]

i'll test it tonight Gary and report the results back here, thanks

[km]

Excellent play Gary, below are the results i got: ROI +11%
The ROI doesn't quite beat Mark's play,but i bet you could tweak it with 1300+ plays to work with, try LEARN ALL for feedback.

ALL 01313 00243 19% 34% 46% 1.11 0.99 0.94 $12.0 0028 $66 1.71

01Feb06 - 31Jan07 MLO 3.0-99.0

Type: Stk Alw Clm Msw Mcl
AgSx: 2yr 3yr 3up 4up Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: ALL
Surf: Turf iTurf

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters: HTR=1 HTR=2
Item Filters:

WK:83-95

[njcurveball]

WOW! You guys are impressive at writing profitable queries!

Thanks for sharing! What an amazing rating HTR is!

Thanks for this challenge Ken. I have a few more things to test now!

[jbshoulda]

MLO 4.0-99.0

Type: Mcl
AgSx: 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: ALL
Surf: ALL

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters: WK=1
Item Filters: WK 85+

TRN:050-350

[km]

Tested it JB, close, lots of bombs with your play; ROI +11% with 1100+

ALL 01120 00154 14% 28% 40% 1.11 0.95 0.90 $16.2 0034 $139 1.29

01Feb06 - 31Jan07 MLO 4.0-99.0

Type: Mcl
AgSx: 2yr 3yr 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: ALL
Surf: ALL

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters: WK=1
Item Filters: WK 85+

TRN:050-350

[Mark]

Well, I was clear for a few panels, but you steeds are flankin' me @ the 1/4 pole and my jock is scrubbin' already...a few more drills in the morning may give me the bottom to run with this pack.

[jbshoulda]

Ken......You forgot to remove the 2 and 3 year olds

[km]

Taking out the 2/3yr made a big difference - JB takes the lead with a 1.21 (+21%) ROI - excellent!

ALL 00817 00114 14% 29% 41% 1.21 0.97 0.91 $17.4 0028 $139 1.31

01Feb06 - 31Jan07 MLO 4.0-99.0

Type: Mcl
AgSx: 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: ALL
Surf: ALL

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters: WK=1
Item Filters: WK 85+

TRN:050-350

[Huguenot]

Hey, after reading this thread I may never have to work again!
:D

I'm wrapping up a writing assignment today -- when that's done I'll tackle this sucker.

[MikeDee]

Ken I have one but it may not qualify since you may not be able to test it.

nfld>6
nMLO>3
nFPSX=1
nFTS=0
nVI<29
nLAY<200

R toppac =1

In English

field size greater then 6
morning line odds greater then 3
FR 1 rank = 1 and is a strong fr1 (nFPSX=1). This is based on PL5.
all races with any first time starters are excluded
VI less then 29
layoff less then 200 days

R toppac =1
The hx4 export has the top pac figure for each horse that is found in the last 10 pps. This is simply the no 1 ranked of the toppacs, excluding any horses that do not have a top pac figure.

[mr. posh]

Km,

Here is my play. Only ran from end of Nov to now. ml 3.0-99.0

PL 3
Clm-MSW-MCL
3u-4u-f-m
10k<>
vi-10-25
sprint
d-wd-id-a
k1-9
AQU-FG-GG-GP-LRL-OP-SA-TAM-TP
FILTER-FC1

plays 233-53-23%-1.51-1.10-1.11-13.3-10-44.0-2.36

[Ritz]

Taking out the 2/3yr made a big difference - JB takes the lead with a 1.21 (+21%) ROI - excellent!

ALL 00817 00114 14% 29% 41% 1.21 0.97 0.91 $17.4 0028 $139 1.31

01Feb06 - 31Jan07 MLO 4.0-99.0

Type: Mcl
AgSx: 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: ALL
Surf: ALL

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters: WK=1
Item Filters: WK 85+

TRN:050-350

How many plays in this spot & why threshold TRN @ 350 JB?

[Mark]

817 plays in JB's test, it's the first # listed after "ALL".

[km]

MikeDee and Posh

I will test your plays tonight. And i'll get Mike's correct this time.

Further discussion on the nFPSX refer to the Access forum. Sorry, i tried to clean up this thread and forward the posts related to nFPSX, but ended up deleting them all instead. I'm too stressed out :o

[MVM]

Can final tote odds be used as a criteria in these spot plays?

I wasn't certain, as the actual odds are not known until after the race.

And I am also assuming that these plays need to be testable through the robot?

[km]

Assumption correct MVM, no tote odds as we can't extract that ahead of time to reveal the spot play. You can use any factors or rankings found in HTR to make your play, export or robot, but not anything that requires separate programming or complex query that the rest of us can't duplicate easily.

[Mark]

Ken, you can ignore my last spot as it utilized "tote" odds, back to the drawing board.

I deleted the post.

[km]

MikeDee's play does not qualify for the contest because he programmed a ranking of the nTOPPAC from HX4 which cannot be duplicated by most users. I coded a similar rank within my own programmable Robot for this test.

The play also utilizes an obscure HX4 item nFPSX = 1 which locates fps dominance (!FR1), but contains a minor flaw (see Access forum). But that isn't his fault, he used the nFPSX=1 directly from the export as is, so i ran his test under that same parameter. Sometimes a little quirk releases a major discovery ==>

ALL 00946 00163 17% 30% 40% 1.19 1.00 0.89 $13.8 0023 $137

A surprising (+19%) ROI with nearly 1000 plays and 17% winners. The Place stats are excellent as well. Well-done creative thinking Dee! Thx for sharing.

[km]

Posh, your play was restricted to a few major tracks (looks like a tournament play) but the rules specify that the contest requires ALL tracks to be tested. So that's how it tested it - with ALL tracks.

Got a low ROI, but with 5000+ qualifiers, this one can be tweaked and show a profit. In fact, i think i could combine a couple of filters and win the contest with it. HINT: Run it through LEARN ALL.

ALL 05280 00795 15% 29% 42% 0.83 0.86 0.90 $11.0 0090 $89

Races Tested: 5193 01Feb06 - 31Jan07

Type: Clm Msw Mcl
AgSx: 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 PL-3

Dist: 4.5 5.0 5.5 6.0 6.5 7.0 7.5
Surf: DirtFast Wet Artif iDirt

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters: FC=1
Item Filters:

[mr. posh]

Okay, Try this one

PL-3
ml-3-40
clm
males
3/4U
10K<>
vi-10-25
sprint
d-wd-id-a
k1-7
FC1
no state breds
k under 109

[Mark]

A relaxing weekend I hope, Mr. Massa.

I've another submission at your leisure:

MLO: 6.0 - 20.0

Type: Clm Msw
AgSx: 3up 4up Fems Male
Misc: Purse<$10k Purse=>10k Vi All (10-50) NO State-Breds PL-5

Dist: 5.5 - 9.5
Surf: ALL

Krnk: K1 K2 K3

Rank Filters: SP=1 SP=2

LAY: 002-030 EPR:080-099 TRN:200-550 JKY:200-550 PED:350-990

Thank you.

[km]

Posh and Mark, will do on your tests, have it by Monday morning - thanks

[mr. posh]

KM Here is my #2

PL-3
Stk/alw-msw-mcl
3
m
Purse 10k>=
vi 33-37
no state breds
Dist=all
d-wd-id-t-it
k1-6
ped 1/2
2ts

[km]

Posh #1

Just 0.92 Bill, where did i go wrong ? (review specs below)

ALL 02005 00313 16% 29% 43% 0.92 0.87 0.93 $11.8 0043 $89 1.53

01Feb06 - 31Jan07 MLO 3.0-40.0

Type: Clm
AgSx: 3up 4up Male
Misc: Purse<$10k Purse>=$10k Vi10-25 NO State-Breds PL-3

Dist: 4.5 5.0 5.5 6.0 6.5 7.0 7.5
Surf: DirtFast Wet Artif iDirt

Krnk: K1 K2 K3 K4 K5 K6 K7

TRKS: ALL

Rank Filters: FC=1
Item Filters:

(K):050-108

[km]

Mark #2
You did it Mark - took the lead! and with a K =1,2,3 play - well done
ROI = 1.25, 21% winners, 763 plays in a year

ALL 00763 00162 21% 34% 49% 1.25 0.96 0.92 $11.8 0020 $59 1.78

01Feb06 - 31Jan07 MLO 6.0-20.0

Type: Clm Msw
AgSx: 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 NO State-Breds PL-5

Dist: 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5
Surf: ALL

Krnk: K1 K2 K3

TRKS: ALL

Rank Filters: S/P=1 S/P=2
Item Filters:

LAY: 002-030 EPR:080-099 TRN:200-550 JKY:200-550 PED:350-990

[centurius]

Type: All
AgSx: 3yr , 3up 4up Fems Male
Misc: Purse>=$10k Vi=all NO State-Breds PL-5

Dist: 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10
Surf: NO ARTIFICIAL

Krnk: K1 K2 K3 K4 K6

TRKS: ALL

Rank Filters: FR2=1 PAC=1
Item Filters:RS=E RS=P RS=S RS=R, $$

TRN:205-550 JKY:205-550 PED:260-990

[km]

Centauris play; ROI = 1.23 (+23%).
But the play count (590) fell just below the contest minimum of 600.
Thanks for submitting the play Centauris, check my specs below to make sure i got it right.

ALL 00590 00111 19% 33% 46% 1.23 1.02 0.98 $13.0 0013 $74 1.48

01Feb06 - 31Jan07

Type: Stk Alw Clm Msw Mcl
AgSx: 3yr 3up 4up Fems Male
Misc: Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 NO State-Breds PL-5

Dist: 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0-14.0
Surf: DirtFast Turf Wet iDirt iTurf

Krnk: K1 K2 K3 K4 K6

TRKS: ALL

Rank Filters: FR2=1 PAC=1
Item Filters: RS=E RS=P RS=S RS=R $$

TRN:205-550 JKY:205-550 PED:260-990

[MikeDee]

I noted this AM that this thread has over 1100 "views". Apparently there are quite a few more people interested in these spot plays then there are HTR members.

[Mark]

That's funny, I noticed that as well. Should all these plays go forward at depressed prices, we'll have to resort to a coded Pig Latin.

[mr. posh]

Perhaps this site should be for members only.

[mr. posh]

KM what happened with Posh #2

[km]

Posh, i apologize, i did run your test, but obviously forgot to cut and past the result:o Will set it up again, run the result and post it here asap.

Mike and Mark = this is probably the most successful thread in the history of this bbs. While there have only been a few contest players, we have thousands of hits and that tally you see is just the registered users - we get several times that many with non-registered viewers.

As i have mentioned to many of you in emails, i wish i had used 1000 plays as the minimum. That would have reduced the ROI and the backfit nature of the contest. But it would also make it less exciting. It's possible to get 1.50+ ROI with 600 plays - so a little fireworks is what we really wanted here in terms of exposure to make the contest more interesting and fun to watch.

[km]

Posh, i apologize, i did run your test, but obviously forgot to cut and past the result:o Will set it up again, run the result and post it here asap.

Mike and Mark = this is probably the most successful thread in the history of this bbs. While there have only been a few contest players, we have thousands of hits and that tally you see is just the registered users - we get several times that many with non-registered viewers.

As i have mentioned to many of you in emails, i wish i had used 1000 plays as the minimum. That would have reduced the ROI and the backfit nature of the contest. But it would also make it less exciting. It's possible to get 1.50+ ROI with 600 plays - so a little fireworks is what we really wanted here in terms of exposure to make the contest more interesting and fun to watch.

[km]

Posh #2
Great ROI (1.34), way short of plays (075) - did i get the parameters (below) correct Bill ?
The Vi is very restrictive here, was that correct?

Give POSH a lot of credit for configuring this play.
The 3yr Male 2TS is a rich source of overlays.

ALL 00075 00022 29% 43% 55% 1.34 0.96 0.91 $9.1 0001 $38 2.32

01Feb06 - 31Jan07

Type: Stk Alw Msw Mcl
AgSx: 3yr Male
Misc: Purse>=$10k Vi33-37 NO State-Breds PL-3

Dist: ALL
Surf: DirtFast Turf Wet iDirt iTurf

Krnk: K1 K2 K3 K4 K5 K6

TRKS: ALL

Rank Filters: PED=1 PED=2
Item Filters: 2TS

[ERNIE LOGSDON]

don't worry about the roi based on the number of views....I must check/view this one at least 10 times a day( maybe more when I'm bored....or forget that I looked last hour )
after the contest is over I plan to post a couple of plays that were not good for the contest but that I feel have merit in the hope of stimulating thought and get plays posted from others.

IMO: There are not enough subscribers (nice people that they are ) to change any odds,just look at the pool size, do the math on what it would take to change/drop the odds even one point. most of the ratings can not be found unless you're using HTR2.......That's MHO <E>

[mr. posh]

KM I did a little tweaking. I forgot your database will have 2yr olds while mine doesn't, so lets revise to the following:
PL-3
M-Mcl
d-id-wd
sprint
k1-9
vi-all
p-10k<>
2-3yr
ped 450+
ped1/2
2ts
no state breds

[MikeDee]

Erine, I'm sure a lot of the views are "us guys", but I only go into a thread when there is something new to view. It's still a lot of views.

The other thing is that we are using a lot of HTR generated numbers in the spots so it would be difficult for non htr players to duplicate until after they join:)

[km]

Posh #3 I tested both Males and Females and Msw and Mcl, from your specs you may have wanted different - but according to the LEARN ALL, it wouldn't matter, the ROI was about the same and < 0.90 in all cases.

ALL 00766 00136 18% 33% 45% 0.88 0.79 0.76 $9.9 0014 $112 1.53

01Feb06 - 31Jan07

Type: Msw Mcl
AgSx: 2yr 3yr Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 NO State-Breds PL-3

Dist: 4.5 5.0 5.5 6.0 6.5 7.0 7.5
Surf: DirtFast Wet iDirt

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters: PED=1 PED=2
Item Filters: PED 450+ 2TS

[centurius]

Type: Stk Alw Clm Msw Mcl
AgSx: 3yr 3up 4up Fems Male
Misc: Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0-14.0
Surf: DirtFast Turf Wet iDirt iTurf

Krnk: K1 K2 K3

TRKS: All

Rank Filters: A/P=1 VEL=1 PER=1 IMP=1 HTR=1
Item Filters: RS=F RS=E RS=P RS=S Fr3>50 Jky Chg(+)

LAY: 010-090 EPR:080-100 TRN:200-550 JKY:200-550

[km]

Centurius #2
Was expecting a much lower play count with the restrictive nature of the filters, but got 1200+, however the ROI was low (0.94). Good win% though. Check my specs below.

ALL 01202 00445 37% 56% 72% 0.94 0.90 0.92 $5.1 0004 $27 2.81

01Feb06 - 31Jan07

Type: Stk Alw Clm Msw Mcl
AgSx: 3yr 3up 4up Fems Male
Misc: Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0-14.0
Surf: DirtFast Turf Wet iDirt iTurf

Krnk: K1 K2 K3

TRKS: ALL

Rank Filters: A/P=1 VEL=1 PER=1 IMP=1 HTR=1
Item Filters: RS=F RS=E RS=P RS=S Fr3>50 Jky Chg(+)

LAY: 010-090 EPR:080-100 TRN:200-550 JKY:200-550

[DanG]

Ken… (I sense you have a free 30 seconds, so I thought I would fill the void.)

All based on HTR – HX4 exports, Paceline-5

Field Criteria
1) nCHG <>2
2) nFPSX <>1
3) nMLO <25
4) nVI <43
5) nHAG <5
6) nRS >0 and <5
7) nFLD >4 and <10
8) nDIST 5.5f
9) nLFST = 2
10) nSEX = 1


• <> “Not equal to”
• < “Less than and not including”
• > “Greater than and not including”

Plain “HTR” English…

1. No Blinkers on
2. (as per the Mike Dee discussion on FPSX) “Not” FR1
3. Morning line less than 25-1
4. VI less than 43
5. Horses age less than 5yo.
6. Running Style not “0” and not 5, or Rear.
7. Field Size greater than 4 and less than 10.
8. Today’s Dist = 5.5f
9. # Lifetime starts =2, (Or Third time starter)
10. Pure discrimination with Males only

Thanks Ken, this is a blast! :)

BTW: On the preview the formatting is all whacked out, sorry about that. :confused:

[centurius]

Type: Stk Alw Clm Msw Mcl
AgSx: 2yr 3yr 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: ALL
Surf: DirtFast Turf Wet iDirt iTurf

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS:ALL

Rank Filters: none
Item Filters: bx

WK:80-95

[km]

DanG play got a 1.35 ROI (+35%) and he takes the lead !
That $225 winner helped the cause ;)
Hey, this is a contest, not a science exam, get it any way you can. No cap.

Nice going Dan - you must have spent a long night retro-fitting this play in Access, yes?

ALL 00658 00125 19% 34% 46% 1.35 1.01 0.84 $14.2 0020 $225

Today’s Dist = 5.5f
Lifetime starts =2, (aka: Third time starter or "3TS")
Males only
NOT Blinkers on
NOT FR1!
MLO < 25-1
VI < 43
Horse's age < 5
Running Style F, E, P, S
Field Size > 4 and < 10 horses.

[DanG]

you must have spent a long night retro-fitting this play in Access, yes?

Ken; (Feel free to delete this; I know this thread doesn’t need any editorials. The plays themselves will more than fill it up.)

However, for the record…Would you believe 30 minutes tops in real time thanks to the HTR robot?

This contest (sparked by Cliff’s thread) has started many discussions concerning HTR, its Spot Play and general data mining capabilities. A clue for me when someone discusses HTR and claims how “limited” the robot is shows me they…a) haven’t tried it, b) used it for 30 minutes and threw up there hands.

To show just how misinformation can be spread there is a thread on PA where someone claims (who says they have used HTR) that HTR’s data mining can’t compare to what’s achievable using Excel or Access…Huh? :confused:

I call one computer I have my “crime laboratory”. This machine runs your Robot and any long run reports I use with HTR.

After “forensics” is done drawing me a map, I use the Robot’s “Learn All” report drill down and give me the clues where to fish in Access.

Your right…I would be burning the midnight oil starting from scratch in Access. HTR however is built for the tournament player, the straight player and certainly the spot player.

For what its worth: Spot Plays as “Mutual Funds”

Sidebar: I never played spot plays in my life until several months after subscribing. I now have a system setup that runs about 35 plays a day of various methods. Like any good mutual fund I’ve found that diversification is key to balance the roller coaster seasonal effects each individual fund will experience. I highly recommend anyone attempting this build a portfolio of a wide range of funds.

• Blue Chips: Strong win% and potential rebate plays. (If you can get them…See Ian soon ;) )
• Growth Funds: The bomb seekers who have long droughts and need feeder capital from you blue chips etc.
• Income Funds: These are your absolute best plays. I only have three, but they are consistent earners and must remain capitalized.

Thanks for the forum Ken and now in 30 seconds this message will self destruct. :D

[njcurveball]

Great post and great job Dan!

You set the bar rather high for the rest of us now! 35% profit on 600+ plays is enough to earn a nice living! And a solid 19% winning percentage to keep profits churning.

Love this thread Ken! Thanks for your generoisty in getting it started. I am sure it will be used for research for many years to come!

[mr. posh]

KM,
This is the greatest newsletter of all time.

[Huguenot]

Darn you Dan! I had a promising turf spot play with an ROI in the $1.20s :(

In the words of one of my favorite movies, "I'll get you my pretty and your little dog, too!"

(of course, "pretty" is not meant to be taken literally)

:D :D

[mr. posh]

KM,
For those of us who don't use access how do we check Dan's play. (not blinkerrs on, not fr1, under 5 yrs old. vi < 43").

[DanG]

KM,
For those of us who don't use access how do we check Dan's play. (not blinkerrs on, not fr1, under 5 yrs old. vi < 43").
Mr. Posh;

A few of those variables are only available via the export. (FPSX, exact horses age etc.) The geneses of the play (3TS) is directly from the Robot’s learn all function.

Truthfully HTR’s robot hits all the key points and has a much stronger probability of going forward than tweaking a back fit with Access.

[DanG]

Darn you Dan! I had a promising turf spot play with an ROI in the $1.20s :(

In the words of one of my favorite movies, "I'll get you my pretty and your little dog, too!"

(of course, "pretty" is not meant to be taken literally)

:D :D
LOL… :D

Long way to go Hug.

Were only on the first turn in a 10f classic.

[qhrick]

MLO 3.5 & UP
Dist:: 8.0 8.2 8.3 9.0 9.5
Krank: K3 K4 K5 K6 K7 K8 K9
HTR=1
WK 85+
Trn: 050-400
Jky: 125-390

default Robot settings for all others

[km]

Centurius #2 Play

Blinkers OFF is always a good producer, but this one did not crack 1.00 ROI

ALL 01152 00188 16% 30% 44% 0.98 0.99 0.92 $12.0 0025 $80 1.43

01Feb06 - 31Jan07

Type: Stk Alw Clm Msw Mcl
AgSx: 2yr 3yr 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: ALL
Surf: DirtFast Turf Wet iDirt iTurf

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters:
Item Filters: bx

WK:80-95

[jbshoulda]

MLO 6.0-99.0

Type: Stk Alw Clm Mcl
AgSx: 3yr 3up 4up Fems Male
Misc: Purse>=$10k Vi10-25 Vi26-32 PL-5

Dist: ALL
Surf: DirtFast Turf Wet iDirt iTurf

Krnk: K1 K2 K3 K4 K5 K6 K7

TRKS: ALL

Rank Filters:
Item Filters: RS=F RS=E RS=P RS=S bx

[km]

The bar is rising in this contest!

qhrick has done it - outstanding 1.38 ROI (+38%) with 700 plays and 19% winners
And it was a rather simple set of Robot parameters, exploiting HTR=1 and Wk 85+.
Well done Rick from the OC!

ALL 00700 00136 19% 33% 46% 1.38 1.05 1.00 $14.2 0026 $86 1.65

01Feb06 - 31Jan07 MLO 3.5-99.0

Type: Stk Alw Clm Msw Mcl
AgSx: 2yr 3yr 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: 8.0 8.5 9.0 9.5
Surf: ALL

Krnk: K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters: HTR=1
Item Filters: WK 85+

TRN:050-400 JKY:125-390

[km]

QHrick's lead didn't last long - JBShoulda has tied his ROI = 1.38 (+38%) but got more plays (745) so he takes the lead - super job JB, another simple Robot structure with Blinkers Off as the key element.

Still two weeks to go in this contest, betcha we hit ROI > 1.50 by the end.

ALL 00745 00111 15% 25% 37% 1.38 1.04 0.98 $18.5 0034 $168 1.39

01Feb06 - 31Jan07 MLO 6.0-99.0

Type: Stk Alw Clm Mcl
AgSx: 2yr 3yr 3up 4up Fems Male
Misc: Purse>=$10k Vi10-25 Vi26-32 PL-5

Dist: ALL
Surf: DirtFast Turf Wet iDirt iTurf

Krnk: K1 K2 K3 K4 K5 K6 K7

TRKS: ALL

Rank Filters:
Item Filters: RS=F RS=E RS=P RS=S bx

[centurius]

Type: Stk Alw Clm Msw Mcl
AgSx: 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: 5.5 6.0 6.5
Surf: DirtFast

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters:
Item Filters: $ $$ Tr Chg (c)

[jbshoulda]

Ken I think you forgot to remove the 2 yr olds from the list. If you don't mind running it again.

[km]

My bad JB, removing the 2yrs improves ROI = 1.42 (+42%) - excellent! You're the leader.

ALL 00708 00106 15% 25% 37% 1.42 1.06 0.97 $18.9 0033 $168 1.39

01Feb06 - 31Jan07 MLO 6.0-99.0

Type: Stk Alw Clm Mcl
AgSx: 3yr 3up 4up Fems Male
Misc: Purse>=$10k Vi10-25 Vi26-32 PL-5

Dist: ALL
Surf: DirtFast Turf Wet iDirt iTurf

Krnk: K1 K2 K3 K4 K5 K6 K7

TRKS: ALL

Rank Filters:
Item Filters: RS=F RS=E RS=P RS=S bx

[njcurveball]

WOW! 40%+ ROI and there is over 2 weeks left!

I don't have all the tracks in my database, but I guess that means it is time to get in the running.

Let's try this one!

Using PL-5

Type : Mcl
Sex : Females
Age : All
Dist: All
Surf: All
EP rank = 1
LAYoff days < 61
RS not equal 'F' or 'R' , all others ok

Now the funky stuff


TRN rating < 220
trainer rank > 2
jockey rank > 1
FX rank > 1
LP rank > 2
Class rank > 1

(counter logic parameters add value!)

OK Ken, let er rip!

[Mark]

Mark #3:

MLO 8.0 - 20.0

Clm
3up 4up Fems Male
Purse <=> $10k Vi All Cla = 1 PL-5

Dist: ALL
Surf: DirtFast Artif iDirt

K1 - K8

HTR = 1
$$

LAY: 012 - 365 PED: 350 - 990

[Pizzaman]

nWKSC>=84
rMLO>=4
nWgtCh>=-2
rEP=1
rHTR=1
nAGE=1

[km]

Centurius #3 ROI = 1.02

ALL 01196 00150 13% 23% 34% 1.02 0.88 0.81 $16.3 0042 $65 1.05

01Feb06 - 31Jan07

Type: Stk Alw Clm Msw Mcl
AgSx: 3yr 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: 5.5 6.0 6.5
Surf: DirtFast

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters:
Item Filters: $ $$ Tr Chg (c)

[km]

NJ's play - great ROI (+26%) but not enough plays (429) to qualify for contest prize

ALL 00429 00076 18% 32% 49% 1.26 0.99 1.01 $14.2 0014 $97

01Feb06 - 31Jan07

See NJ's spec list above, this one has to programmed or queried in a db from HX4 output.

Basic play is E/P = 1 in Female Maiden Claimers, with a number of interesting filters (!)

thanks for participating Jim

[km]

Mark #3

Mark, you have done it - 1.50 ROI (+50%) - brilliant!

When i first looked at this one i thought it was very interesting, combining Class=1 and a PED filter with Claiming races only. A simple but clever Robot play. Well done and ROI profit across the board on Place and Show too.

ALL 00769 00139 18% 30% 43% 1.50 1.12 1.04 $16.6 0032 $111 1.56

01Feb06 - 31Jan07 MLO 8.0-99.0

Type: Clm
AgSx: 3yr 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50

Cla=1

PL-5

Dist: ALL
Surf: DirtFast Artif iDirt

Krnk: K1 K2 K3 K4 K5 K6 K7 K8

TRKS: ALL

Rank Filters: HTR=1
Item Filters: $$

LAY: 012-365 PED:350-990

[km]

Pizzaman Play - good result (ROI = 1.26) with lots of qualifiying horses (926) and decent win%

Another strong return with HTR=1 at the core

ALL 00926 00202 22% 36% 48% 1.21 0.98 0.91 $11.1 0021 $86

nWKSC>=84
rMLO>=4
nWgtCh>=-2
rEP=1
rHTR=1
nAGE=1

[DanG]

Mark #3

Mark, you have done it - 1.50 ROI (+50%) - brilliant!

When i first looked at this one i thought it was very interesting, combining Class=1 and a PED filter with Claiming races only. A simple but clever Robot play. Well done and ROI profit across the board on Place and Show too.

ALL 00769 00139 18% 30% 43% 1.50 1.12 1.04 $16.6 0032 $111 1.56

Mark,

Just outstanding!

Especially considering the tight scope of criteria. Dare I say, this is the type of play that can go forward.

BTW: Lurkers beware…The play looks deceptively simple, yet there are 5 key elements that are exclusive to HTR. (Actually many more than 5 as the HTR consensus covers so many variables.)…Oops…make that 6 elements…Marks brilliant design! :)

[Mark]

Thank you Ken and Dan. It was very self-satisfying to hit the 1.50 target.

That target is now on my back and will surely be exceeded. Regardless, the process forced me to address high mutuels and stop pining over Pinnacle's demise.

[MikeDee]

I'm having trouble duplicating your spot play in access Mark.
In particular coming up the the number of plays

In export language I am using
nmlo >=8 and <=20
nage =1 (3up or 4up)
nsurf <>2 (all but turf)
ncla=2 (claimers only)
rk >=1 and <=8
rhtr=1
rclass=1
nlong=2
nlay >=12 and <=365
nped >=350 and <=990

does this look right?

[DanG]

I'm having trouble duplicating your spot play in access Mark.
In particular coming up the the number of plays

In export language I am using
nmlo >=8 and <=20
nage =1 (3up or 4up)
nsurf <>2 (all but turf)
ncla=2 (claimers only)
rk >=1 and <=8
rhtr=1
rclass=1
nlong=2
nlay >=12 and <=365
nped >=350 and <=990

does this look right?
Mike;

Funny you should say that.

Mark will tell you I emailed him with the same concern. We must be misinterpreting something? I come well short in play count also.

[MikeDee]

I'm not a robot person since I use access. But in access a rhtr=1 AND a rclass=1 AND a nMLO>=8 is pretty restrictive on play count.

Maybe the robot is doing something a tad different since the class 1 and class 2 buttons are not in the filter box. If I run it with out any class button checked I get a number of plays that is in the ball park.

[km]

INQUIRY! Uh oh Mark, they are looking at the head-on.

Remember, that MikeDee is the other steward here, and he has to confirm my test before the purse money can be distributed

My criteria were cut and pasted directly from the Robot and you can see them above.

But you know what?
If it is my mistake, there is still some play with 700+ samples still produced a 1.50 ROI ! Maybe we'll find something by accident anyway.

If you take out the Class = 1, do you get a similar result Dee?

Here are today's (Friday 2/16) plays with Mark's criteria as i plugged them in to Robot and GET MY PLAYS - see if they all match
09:58 AQUi-02 #6 Chicago 15/1
10:41 BEU-02 #6 Kyrburg 10/1

[Mark]

Gentlemen, I've just returned from a celebratory wine purchase to the thumping of blinking lights:

Ken's parameters differ from mine in two instances which I just reviewed after reading your concerns. Ken added 3yr and used MLO 8-99/1 rather than 8-20/1. It seems unlikely those differences would lessen my Robot results...they are from two folders that overlap the contest dates yielding 611 plays @ 1.55 (3-18-06 through 2-11 07) and 79 plays @ 1.63 (1-26-06 through 3-17-06). The total was 690 @ 1.56 ROI and I felt comfortable not deducting the 15 day overlap prior to posting. I have most major and minor tracks and didn't use the "unlisted" Robot designation. Assuming Ken would have more tracks in his ALL, the ROI drop to 1.50 and play increase to 769 didn't raise a flag.

Beyond that, I'm at a loss to explain your results. I'll delay popping a cork until until advised.I'll also rerun the tests and post any differences forthwith.

[Mark]

My phone # is 508-675-9820 if required or preferred.

[Mark]

Mike, your Access specs look exactly as I submitted. As related Ken inadvertantly added 3yr and upped the MLO to 99/1.

I reran my tests, factoring out the extra days, and got 668 plays @ 1.55 ROI. It appears there is either a problem with the Robot or Access.

[jbshoulda]

Mark this is what I got with the dates listed and I do not have all tracks. So it seems that Ken's test is right as it has Feb which is not in my test and this is robot only.


ALL 00511 00093 18% 30% 45% 1.61 1.18 1.13 $17.7 0023 $111 1.61



Races Tested: 12541 22Jul05 - 15Feb07 MLO 8.0-20.0

Type: Clm
AgSx: 3up 4up Fems Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 Cla=1 PL-5

Dist: ALL
Surf: DirtFast Artif iDirt

Krnk: K1 K2 K3 K4 K5 K6 K7 K8

TRKS: ALL

Rank Filters: HTR=1
Item Filters: $$

LAY: 012-365 PED:350-990

[MikeDee]

I agree with your 2 plays as todays plays Ken

[MikeDee]

I do get a similar result when I take out rclass 1.

[ERNIE LOGSDON]

For what ist worth dept.
I am not one of the stewards and don't like watching Inq replay's

my robot test report on 55578 races form 1feb2006 to 31jan2007 reports:

18351 races--700 plays--127 winners--18%W--30%WP--43%ITM--$1.55 WROI--$1.14 PROI
--$1.06 SROI--$17.1 $AVGM--30 $20+ --$111 HIGH --1.58 I.V.


TESTED (below the dotted line at the bottom of the report---don't know what to call each item---)

reacs tested 18351---01feb06 - 31jan07 ---MLO 8.0-20.0 ---TYPE= CLM ---
AGSX= 3UP-4UP-FEMS-MALES---CLS-1 ---PL-5----DIST= ALL---SURF=DIRTFAST ARTIFI iDIRT-------Krnk = 1thru 8 ----tracks =ALL -----RANK FILTERS= HTR=1 ----
ITEM FILTER = $$ ---LAY = 012-365 ----PED= 350-990

---------------------------------------------------------------------------------------
Hope this is helpful, seeing how I can't find one to break 600 and $1.25 ( YET ! )
I have to say very good job.....thanks for the contest and to all who have shared their hard work<Ernie>

[Mark]

Thanks, Ernie. Your results mirror mine.

Have a great night.

[njcurveball]

I do love the ladies! Let's try them again!!!


Using PL-5

Type : Mcl
Sex : Females
Age : All
Dist: All
Surf: All
ML > 5 and ML < 25
EP rank = 1
NQP > 0 (at least 1 Quirin Speed point)
RS not equal 'F' or 'R' , all others ok

NOW we add 2 conditions (first to make sure the race has some contention for the lead to get a price)

NQFIVES > 0

Then to make sure the horse has some "uncertainty"

Last finish > 3 (NLFIN > 3)
and RLP > 3 (pace line too strong)


This is for the Access people

WHERE
nsex = 2
and ncla = 4
and rep = 1
and
nmlo > 5
and nmlo < 25
and nrs > 1 and
nrs < 5 and
nqp > 0
and
rlp > 3
and nlfin > 3
and nqfives > 0


OK Ken, let er rip again!

[km]

Oh my goodness - you have done it Jim! +62% ROI and and a real bomber play.

Tournament bums will like this one and it had 800 hits. Clever play NJ.

Every day we get a new leader. Will this contest end with > 2.00 ROI ?

Play All 00806 00114 14% 26% 37% 1.62 1.20 1.02 $22.9 0044 $144 1.32

01Feb06 - 31Jan07

Type : Mcl
Sex : Females races only
Age : All
Dist: All
Surf: All
ML > 5 and ML < 25
E/P rank = 1
NQP > 0 (at least 1 Quirin Speed point)
RS = E, P, S
QFIVES > 0 (at least one horse in the race has QP >= 5)
Last finish > 3
L/P rank > 3

[Mark]

A smashing success. It feels good to have a target again.

It seems difficult to imagine topping you through the Robot alone. Dig we must.

Congratulations, Jim, a fine weedend is a given.

Regards.

[Donnie]

....you guys are just incredible! You are giving me, a guy who handicaps "ïn the moment", pause for thought!

Can't you just hear all the keyboards setting these up in Access or in the robot as personal plays??:D

[dehere]

Donnie - been there - just done that :)

I get two plays for NJCurveball play for eight tracks for today (GP 2nd and OP 4) - anyone agree? or disagree? None for Mark's play.

tracks used

GP, OP, AQU, BM, SA, FG, LRL, TAM

[Donnie]

Jesus H. Christ, Henry!! Couldn't you at least post the horses that qualified in those races for all the Lurkers?? (and those of us too lazy to have that programmed in yet!!)

BTW, my wife says the "H." stands for Howard. A very Jewish name, I guess. :D

[HBee]

BTW, my wife says the "H." stands for Howard. A very Jewish name, I guess. :D

Donnie, thanks for clearing that up. It's one of those questions you always wondered about, but never borhered to ask :D

Herman

[Zaf]

Donnie - been there - just done that :)

I get two plays for NJCurveball play for eight tracks for today (GP 2nd and OP 4) - anyone agree? or disagree? None for Mark's play.

tracks used

GP, OP, AQU, BM, SA, FG, LRL, TAM

Agree, I get the same for the tracks listed above.


There are some plays for a few of the Mickey Mouse tracks as well.

[njcurveball]

Thanks guys,

You gave me a good idea! The horse in the 4th at Oaklawn has a trainer change, but was not claimed.

#3 On Seattle Time, trainer change from Calhoun to Adams.

OK Ken, at risk of getting stoned here, can I add another condition to my play? I would like to take out the horses that were claimed NTRCH = 2. That would make the play as follows:
Type : Mcl
Sex : Females
Age : All
Dist: All
Surf: All
ML > 5 and ML < 25
EP rank = 1
NQP > 0 (at least 1 Quirin Speed point)
RS not equal 'F' or 'R' , all others ok

NOW we add 2 conditions (first to make sure the race has some contention for the lead to get a price)

NQFIVES > 0

Then to make sure the horse has some "uncertainty"

Last finish > 3 (NLFIN > 3)
and RLP > 3 (pace line too strong)

*** and NTRCH < 2 (no recent claims)

IN Access

WHERE
nsex = 2
and ncla = 4
and rep = 1
and
nmlo > 5
and nmlo < 25
and nrs > 1 and
nrs < 5 and
nqp > 0
and
rlp > 3
and nlfin > 3
and nqfives > 0
and ntrch < 2


*** NOTE: If this will get me stoned, then disregard this message. :eek:

[km]

You already have the lead in the contest NJ, keep that tweak for your ace in the hole in case someone passes you later - then we'll test it :)

[mr. posh]

Again for us po folk! How can this be done in the robot?

Access may be great but allof us don't have it or understand it.

How can it be a winner if everybody can't "access" the results.

I thought it was supposed to be run on robot and then confirmed by access. How can it be the other way around, when several of his qualifiers cqan't be found in robot.

[km]

I agree with what Bill (Posh) is saying. Only 20% of the HTR subscribers use the export + db, virtually everyone that downloads with us has used the Robot. So your point is well taken, the purpose of this contest was to showcase HTR and the many ways of making money with it - which we have sure proven can be done with a db or the Robot, but it would be better if everyone can duplicate the play.

However - the original rules did not specify any particular research tool. They allowed any spot play that I could re- test. So in terms of the contest right now, NJ's play continues to be the leader and he will win the $500 if his 1.62 ROI holds up.

We'll have another contest later that requires the user to utilize the parameters found in the Robot only and will spell out the rules more clearly. You can still use a db query, but the criteria will have to be available in the 'bot.

NJ's play can be emulated somewhat in the Robot, see my post on thread for "Contest LEader"

[MikeDee]

NJ

In looking at your tweak You mentioned that you made 2 changes but I only see 1 listed in the access version and that is

ntrch < 2

Was there only one? or is there another?

[njcurveball]

Sorry Mike,

I cut and paste the previous query, so adding NTRCH < 2 would be the only tweak. I actually agree with Mr. Posh on the "conditions".

But I also would suggest last race finish, quirin speed points, and the Qfives in a race be added at some point. I tried to make my query logical from a handicapping standpoint. Those factors do make sense. I hope some would agree that these parameters actually are less "tweakable" than things like PED, WK rating and VI. Not saying those are not great handicapping factors, just that there is a much wider range to play with.

I also would agree that with a database you could backfit a lot finer. I probably could get a few queries over $2 if I used some crazy stuff.

This contest certainly has stimulated ideas and I would love to see that continue. If you guys feel using a DB is against the "spirit" of the contest, I will gladly remove my entry. The ideas on this thread are worth much more than the prize.

Jim

[AwolAtHTR]

Ken said:

Spot Play Contest!

contest open to any registered bbs user using HTR Export or Robot on HDW data files



and Jim, as the leader offered:

This contest certainly has stimulated ideas and I would love to see that continue. If you guys feel using a DB is against the "spirit" of the contest, I will gladly remove my entry. The ideas on this thread are worth much more than the prize.


However, those that make the rules can change the rules!!

I am an Access user but find using the Robot is great for generating rules and clues for a spot play. So, the change MIGHT decrease my chances of finding that 1.88 ROI to win the contest. However, with a Robot only rule, the issue about validation is simplified and a corrupted data base will not be an issue.

Also, accept the Robot result as reported by Ken. That is, Ken said Mark's was 1.50 but Ernie reported 1.55. I believe, that three HTR users with the same data period could report THREE different (but close) results as a result of how each users keeps their data clean.

With this change, we will have Mark's 15x as the target and I am betting that the final ROI will be closer to 1.88 then 1.50 ROI.


so, I vote for Ken to accept Jim's offer and restrict entries by ONLY the Robot.


duane

[km]

AWOL makes a good case. And NJ Jim is a good sport.

So let's do it democratically - i'll put up a poll and we'll abide by the majority.
(see Poll thread in this forum and cast your vote)

[km]

Status quo on the rules - the majority have spoken - any criteria that i can test from HTR is acceptable - NJ continues to lead.

[Mark]

Mark #4

MLO 8.0 - 20.0

Clm
3up 4up Male

Purse: ALL
Vi: ALL

Cla= 1 PL - 5

Dist: 5.5 - 9.5
Surf: DirtFast Artif iDirt

K1 - K8

$$
N0 FIG2 = 1 (Ken, I don't know this coding)

LAY: 012 - 365 PED: 350 -990

[km]

Mark #4

Mark, you have done it again! 1.64 ROI with 680 plays. Wow.

Hold off on the celebration while i test NJ's tweak to his 1.62 play (nTRCH < 2).
Will post in the next thread.

Note that Mark #4 requires Export HX4 field 101
rPFIG <> 1
but the rest of it can be done in the Robot as below

ALL 00680 00108 16% 26% 37% 1.64 1.12 0.97 $20.6 0039 $111 1.38

01Feb06 - 31Jan07 MLO 8.0-20.0

Type: Clm
AgSx: 3up 4up Male
Misc: Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 Cla=1 PL-5

Dist: 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5
Surf: DirtFast Artif iDirt

Krnk: K1 K2 K3 K4 K5 K6 K7 K8

TRKS: ALL

Rank Filters:
Item Filters: $$

LAY: 012-365 PED:350-990

Additional: Not FIG2=1 (nPROJ <> 1 Field 101, HX4)

[km]

Applied NJ's "no claims" tweak to his previous play, and now it gets = 1.63 with 797 plays.
Missing Mark by .01 !!

The tweak cut a few losers, but not quite enough to get the ROI past 1.64. But with 797plays to work with, i bet he'll squeeze that extra 2 cents for the ROI. Go for it NJ.

Play All 00797 00112 14% 26% 37% 1.63 1.21 1.02 $23.2 0044 $144 1.32

01Feb06 - 31Jan07

Females
Mcl
E/P = 1
MLO 5-25
RS: E, P, S
QP > 0
L/P > 3
LFIN > 3
QFIVES > 0
NOT Claimed Last

[njcurveball]

Thanks for testing Ken.

And congrats Mark! Way to go!

[qhrick]

quick question

are final odds valid for this contest?

[Rick]

That one has already been answered.

NO

That is something that is only known after the race was run.

[qhrick]

That one has already been answered.

NO

That is something that is only known after the race was run.

oops!!!!!!!!!!!!

[DanG]

Mark #4
Mark, you have done it again! 1.64 ROI with 680 plays. Wow.

Time out!

I ran Marks play through Access and I only got 13 plays total? :confused:

Just kidding!!! :D :D :D

Great play Mark!!!

The cooler weather must agree with your ‘grey matter. Again, what power in HTR to achieve a 64% gain with a very creative play that is so focused.

[Mark]

Thank you gentlemen. Dan, your humor wasn't lost on me after I righted myself and took another nitroglycerin tablet.

[njcurveball]

Well here we go again!

Ken can you please add the condition NWKSC > 66.


thanks,
Jim



01Feb06 - 31Jan07

Females
Mcl
E/P = 1
MLO 5-25
RS: E, P, S
QP > 0
L/P > 3
LFIN > 3
QFIVES > 0
*** NWKSC > 66

[km]

Jim, your specs above do not include nTRCH < 2, did you still want that item or have you eliminated it for this particular play?

[njcurveball]

Ken,

Please do not include them, the # of horses claimed is too small to be a criteria for me.



thanks,
Jim

[njcurveball]

One more thing Ken,

I want to include horses with NWKSC = 66.01, 66.02, etc. So if I make the bottom number in the robot 66 I am hoping that will include them.

So perhaps this is a match if I say NWKSC > 65.99?

thanks,
Jim

[centurius]

Type: Stk Alw Clm Msw Mcl
AgSx: 3yr 3up 4up Fems Male
Misc: Purse>=$10k Vi10-25 Vi26-32 Vi33-37 PL-5

Dist: 4.5 5.0 5.5 6.0 6.5 7.0 7.5
Surf: DirtFast iDirt

Krnk: K1 K2 K3 K4 K5 K9

TRKS: ALL

Rank Filters:
Item Filters: Fr3>50 PED 450+ J+T>30%

LAY: 002-090 EPR:080-100 TRN:200-550 JKY:200-550 WK:65-84

[km]

NJ's just toying with us (+72%) !! Masterful.

Latest tweak
ROI = 1.72
Plays = 694

Play All 00694 00100 14% 26% 37% 1.72 1.20 1.03 $23.9 0041 $144 1.35

Females
Mcl
E/P = 1
MLO 5-25
RS: E, P, S
QP > 0
L/P > 3
LFIN > 3
QFIVES > 0
WK > 66

I programmed it as you stated in the specs:
nWK > 66, same as nWK >= 66.0001

Robot users would set it as
WK 66-95
which would include the value 66.0000, but this will have no real effect on the ROI.

NJ and Mark have probably squeezed as much water out the sponge as they can with their respective plays - but the bar has been set so high - can someone exceed it with 10 days left in the contest?

[Mark]

What can I say Jim, my day in the catbird seat was worth the effort.

"Hats off" to you and your skills.

[DanG]

Fantastic Jim!!!

Jim is doing his P. Val impersonation and opening up lengths on the far turn. By next weekend everyone will be going to the mattresses ala Clemenza in Godfather-I. ;) :D

[Donnie]

nDIST>=8
nSEX=1
nPURS>=10
nJKCH 1 or 2
rHTR=1
nMLO between 6 and 12 (inclusive)
nPED>500


As Jim would say....Let 'er rip!!
(don't know how it will do over a year, but with just a couple days missing from this year, shows me 105% ROI for WIN, 22.5% PLA, 14.6% Show! Hopefully it's not short on number of plays!!)

[njcurveball]

Jim is doing his P. Val impersonation


I am going to the drugstore to get some Nair and a few razors in case I win and Ken comes to test me. :D

[km]

Donnie's play

Very restrictive parameters, i got 1.17 ROI with just 143 plays

Where did i go wrong Donnie?
check my criteria below

ALL 00143 00026 18% 29% 44% 1.17 0.86 0.94 $12.9 0004 $42 1.52

01Feb06 - 31Jan07

MLO 6.0-12.0

Type: Stk Alw Clm Msw Mcl
AgSx: 2yr 3yr 3up 4up

Males

Misc: Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: 8.0 8.5 9.0 9.5 10.0-14.0
Surf: ALL

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters: HTR=1
Item Filters: Jky Chg(+)

PED:500-990

[qhrick]

nsurf<>2
Rk 3-8 ( no k1, k2, k9 )
htr=1
jock 105-270
trainer <=300
Nvi<=32
distance 6.5 and all 8.0 &up
Nmlo>=6
nepr 83-95
nqp<>1 ( no quirin points = 1 ) less than 1 ok tho

[Donnie]

KM-
You have the Jockey as +...is that both the 1 and the 2?

It may be that.....as of the first of the year I got 36 plays with a week missing in my db....was hoping it would "carry-out"....check to see if you have both Jockey changes accounted for.....but I don't think it is gonna carry me to 600 plays! Back to the salt mines tonight then! But YTD...was 105%.

[DanG]

DanG #2

Alright, let’s try this again. Well bred long shot males are the theme. Let’s go for lucky 13 in the criteria. :)

• rPED = 1 (Quality parents)
• nSEX = 1 (Excluding the ladies)
• nEPR >97 and <111 (not including)
• nMLO <>15
• nRS <>4 (Close to the pace, or benny from chaos)
• nLFST >5 (Need some racing experience)
• nLONG >0 ($, $$)
• nDS <>5 (No artificial sprints)
• nQP >1 and <8 (not including)
• nFPSX <>3 (Not a dominate FR3, which hurt prices)
• nDist >5.5 (not including)
• nFLD >5 and <12 (not including)
• rHTR <>5 (Like so many subsets, they needed to be either excel, or be ugly)

[km]

Donnie's play - corrected and used nJKYCh > 0 (any jockey change)

Not enough plays to qualify for contest (< 600)

Play All 00310 00048 15% 26% 41% 0.99 0.80 0.87 $12.7 0006 $53 1.32

01Feb06 - 31Jan07

MLO 6.0-12.0

Type: Stk Alw Clm Msw Mcl
AgSx: 2yr 3yr 3up 4up

Male

Misc: Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: 8.0 8.5 9.0 9.5 10.0-14.0
Surf: ALL

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters: HTR=1

PED:500-990
nJKYCH > 0

[km]

qhrick's play

got em Rick +78%, just over the 600 plays - well done!

Play All 00637 00115 18% 29% 40% 1.78 1.25 1.10 $19.8 0046 $86 1.63

nSurf <> 2 (anything but turf)
rK 3-8 ( no k1, k2, k9 )
HTR=1
JKY = 105-270
TRN <=300
VI <=32
Dist = 6.5 and >= 8.0 (no 7.0, 7.5)
MLO >=6
EPR 83-95
QP<>1 ( no quirin points = 1 )

[centurius]

MLO 6.0-99.0

Type: Stk Alw Clm Msw Mcl
AgSx: 2yr 3yr 3up 4up Fems Male
Misc: StateBred Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0-14.0
Surf: ALL

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters:
Item Filters: J+T>30%

[km]

Dan's latest play, i got just 440 hits, what was your total Dan?
If you feel i came up short, i'll double-check my specs, but i followed it exactly as you listed.
Great ROI = 1.73 though.

ALL 00440 00071 16% 27% 38% 1.73 1.04 0.95 $21.5 0026 $114 1.31

HX4 query

rPED = 1
nSEX = 1
nEPR > 97 and <111
nMLO <>15
nRS <>4
nLFST >5
nLONG > 0 ($, $$)
nDS <>5
nQP >1 and <8
nFPSX <>3
nDist >5.5
nFLD >5 and <12
rHTR <>5

[km]

Centurius - i got 5000+ plays and about 0.88 ROI with the simple play you listed
TJ > 30%
MLO > 6/1
was there more criteria??

[centurius]

Did you include StateBred ONLY

[DanG]

Dan's latest play, i got just 440 hits, what was your total Dan?
If you feel i came up short, i'll double-check my specs, but i followed it exactly as you listed.
Great ROI = 1.73 though.


Wow,

Interesting how we could be that far apart. :confused: I have complete results and I get 116 / 623 19% $1.95

Two thoughts: Can Mike run this one and could the “randomness” of nFPSX create that big a discrepancy? (nFPSX total was only a group of only 20 horses eliminated in my PL5 DB)

[MikeDee]

Dan & Ken

With my incomplete data set I get
610 plays
111 wins
18% wins
1.87 ROI

[DanG]

Thanks for running that Mike.

Interesting differences between the export data and not. That would be a great by product of this exercise if the reason was discovered. :confused:

[km]

Cent - i will retest with STATE BREDS only, sorry

Dan - i'm sure i missed something in the specs, got the nFSPX right for sure though because i had it pop up individual horses and checked the screen, but i'll re type my query and try again, it is easy to miss one item and that makes all the difference

thx for testing it 'Dee

[MikeDee]

hey Ken take a quick look at what you set for rhtr. The first time I did it I set it to rhtr = 5 and not the <>5 as is specified. That of course greatly reduced plays.

[njcurveball]

Ken,

Please add the condition ML <> 10 to my ladies.

Females
Mcl
E/P = 1
MLO 5-25
RS: E, P, S
QP > 0
L/P > 3
LFIN > 3
QFIVES > 0
WK > 66
*** MLO <> 10

This has under 600 plays in my database, but I do not have all the tracks. So let er rip.


thanks,
Jim

[km]

Centurius (revised) = -11% (Statebred only)

You can improve this play considerably by using high Wk ratings

ALL 00661 00076 11% 21% 34% 0.89 0.78 0.84 $15.4 0018 $53 1.07

01Feb06 - 31Jan07

MLO 6.0-99.0

Type: Stk Alw Clm Msw Mcl
AgSx: 2yr 3yr 3up 4up Fems Male

Misc: StateBred

Purse<$10k Purse>=$10k Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: ALL
Surf: ALL

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

J+T>30%

[centurius]

I did not have Purse<$10k

thanks

[Donnie]

hey Jim---
can you explain your rational behind the MLO <>10...? Why exclude the 10-1 mlo's only??

[AwolAtHTR]

hey Jim---
can you explain your rational behind the MLO <>10...? Why exclude the 10-1 mlo's only??

see Jim's talk about SNIPER at:

http://www.homebased2.com/forums/showthread.php?t=6292

However, I do not agree about the BackFitting issue which I will post more about in the SNIPER thread. I contend that PariMutuel BackFitting is a classic example of Reverse Engineering.

just let me say here, Reverse Engineering finds an answer according to the history.

When your logic does not agree with the history, maybe you have not seen enough history. The variations on this theme are endless, enuf said for now!!


duane

[Donnie]

Thanks D! Wish I could keep up on these fast moving threads!!

[njcurveball]

Here is one I was working on for the maiden claiming boys. Let's see if it gets enough plays.

NCLA = 4
AND NSEX = 1 (maiden claiming boys)
AND NMLO > 6
AND NMLO < 25
AND NMLO != 10 ( I am finding this sniper has some real value!)
AND RK > 6 AND RK < 9 (bombs away!)
AND NWKSC > 65.99
AND NFTI > 27 AND NFTI < 60 (FTI price range)
AND NCHAOS = 0
AND NVI > 25 ( below this races got unpredictable)


Now I can get to work on the Sports Haven contest. :-)

[km]

Dan #2 Redux +54%; 866 plays

One possibility of disparity between export users and my result is that I'm using the Jan 2007version of HTR2 for all the plays generated. It's possible that many db tables were built from previous versions of HTR2. The $, $$ were considerably revised for 2007.

I coded your specs as shown below and got this result =

ALL 00866 00132 15% 25% 36% 1.54 1.00 0.87 $20.3 0044 $114 1.23

Race filters
PL-5
nSEX = 1 (Males Only)
nEPR > 97 and nEPR < 111
nDS <> 5 (no artificial sprints)
nDIST > 5.5 (distance 6.0f - 14.0f)
nFld > 5 and nFLD < 12 (field size 6 - 11)

Horse Filters
Lifestarts > 5 (6 or more races lifetime)
MLO <> 15 (no 15/1 ML, all others ok)
PED = 1
$ or $$
QP > 1 and QP < 8 (Quirin speed pts 2 - 7)
HTR <> 5 (eliminate HTR = 5, others ok)
nRS <> 4 ( RS = "S" eliminated)
nFPSX <> 3 (not a FR3 = 1! fps differencial > 0.50)

Here is a list of the plays generated from Feb 1 - 22
if you have any that are not the same as these, send them to me (email) and we can find the difference

<racefile><Pn><horse>,<win$>

"0222B04.BM","5","My Captain",0
"0219B01.OP","7","Smokey Ridge",0
"0219B02.OP","6","Mr. Moonshine Time",0
"0219B10.OP","7","Codasco",0
"0219B06.SA","9","Smokin Forest",0
"0218B01.AQU","8","Speedjama",0
"0218B02.GP","9","R. C. Indy",41.2
"0218B08.HOU","1","Victory Bride",0
"0218B04.LRL","2","Leagueofhisown",0
"0218B08.LRL","5","Ironton",15
"0218B10.OP","8","Steel Pier",8
"0218B08.PHA","4","Riverdale",0
"0217B04.FG","1","Tridecagon",0
"0217B10.OP","10","Flashstorm",0
"0217B09.SA","10","Northern Soldier",0
"0216B07.AQU","6","Auto City",34.2
"0216B02.GP","4","Hard Head",0
"0216B03.SA","1","Ocean Grand",0
"0215B04.SA","8","Easy Million",0
"0214B09.GP","9","Tempt Fortune",0
"0213B06.SUN","5","Super Danny",0
"0212B07.PHA","1","Wander Time",0
"0211B09.CT","1","Kaufy Big Shot",0
"0211B08.OP","3","Sir John Q",0
"0211B09.SUN","5","Excessive Contact",0
"0210B09.AQU","10","Mr Sam I Am",0
"0210B02.FG","6","Chachacharlie",0
"0210B08.GG","11","McNasty",14.6
"0210B08.OP","5","Kansas Blues",0
"0209B05.AQU","7","Stopbluffing",0
"0209B09.CT","3","Highly Appealing",38.4
"0209B07.GP","6","Waterville",0
"0208B07.GP","4","Juniper Knight",0
"0208B07.SA","1","On the Acorn (GB)",11.4
"0205B08.FG","8","Group",0
"0203B10.GP","1","Seaside Links",0
"0203B08.TUP","5","Point Dume",0
"0202B10.TP","3","Sir Winzalot",10.8

[km]

Centurius - 3rd test - sorry for previous omissions
-12%, 574 plays

ALL 00574 00067 12% 21% 33% 0.88 0.76 0.83 $15.1 0016 $47 1.10

01Feb06 - 31Jan07

MLO 6.0-99.0

Type: Stk Alw Clm Msw Mcl
AgSx: 2yr 3yr 3up 4up Fems Male

StateBred Only

Purse>=$10k

Vi10-25 Vi26-32 Vi33-37 Vi38-50 PL-5

Dist: 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0-14.0
Surf: ALL

Krnk: K1 K2 K3 K4 K5 K6 K7 K8 K9

TRKS: ALL

Rank Filters:
Item Filters: J+T>30%

[km]

Oh my goodness - NJ's Play gets a whopping +117%, 661 plays, 7% wins

But - HOLD YOUR HORSES - the rules were clear - must exceed 10% win rate to qualify for contest prize - ouch!

Play All 00661 00045 07% 11% 18% 2.17 1.09 0.72 $63.6 0039 $170 0.62

nCLA = 4 (Mcl)
nSEX = 1 (males only)
MLO > 6 and < 25
MLO <> 10 (10/1 ML excluded)
K = 7,8 (Krank = 7 or 8)
WK > 65.99 (Wk rating 66-95)
FTI > 27 AND NFTI < 60
CHAOS = 0 (no chaos races)
VI > 25

Nice backfit for bombs (avg mutuel $63!), but need > 10% wins to qualify for $500 prize.

[MikeDee]

One possibility of disparity between export users and my result is that I'm using the Jan 2007version of HTR2 for all the plays generated. It's possible that many db tables were built from previous versions of HTR2. The $, $$ were considerably revised for 2007.

Now it is making more sense on why past DB exports are not matching up with with your current robot tests. I did not re-run my db with the current versions of HTR2 so any existing figures that have been revised are not in the DB I am testing witn.

This explains why when I ran a test using HTR2 on the same data set of 66 races the Db and the robot got the same results.

I feel better now:)

[njcurveball]

Thanks for testing Ken. I was hoping to get a few wins at the Minor Tracks not in my db.

How did this one make out? Did it have enough plays?

Please add the condition ML <> 10 to my ladies.

Females
Mcl
E/P = 1
MLO 5-25
RS: E, P, S
QP > 0
L/P > 3
LFIN > 3
QFIVES > 0
WK > 66
*** MLO <> 10

[km]

NJ just misses the lead...
his latest tweak = +77%, 736 plays, 16% wins

Play All 00736 00115 16% 27% 39% 1.77 1.21 1.06 $22.7 0044 $144 1.45

Females
Mcl
E/P = 1
MLO 5-25
MLO <> 10
RS: E, P, S
QP > 0
L/P > 3
LFIN > 3
QFIVES > 0
WK > 66

[njcurveball]

OK Ken! That hurt! :(

Please add NFLD > 6

Females
Mcl
E/P = 1
MLO 5-25
MLO <> 10
RS: E, P, S
QP > 0
L/P > 3
LFIN > 3
QFIVES > 0
WK > 66

*** NFLD > 6

thanks,
Jim

[km]

You've done it Jim!

84% profit, 685 plays, 16% winners - pushing the envelope toward the 100% ROI (2.00)!

Play All 00685 00107 16% 26% 38% 1.84 1.22 1.07 $23.5 0043 $144 1.49

Females
Mcl
E/P = 1
MLO 5-25
MLO <> 10
RS: E, P, S
QP > 0
L/P > 3
LFIN > 3
QFIVES > 0
WK > 66
Field Size > 6

[qhrick]

Sure, NJ teases me with his curveball tweak to get head and head, then tosses a NFLD>6 at me to blow me away!!!

Congrats!
-rick

[mr. posh]

Jin,
Can you explain several parts to your successful play.

Why MLO 5-25 and MLO <>10
Also qfives
I think I can look at the rest via robot and then look at what it picks aqnd go from there.

Thanks

[Mark]

Mark #5

MLO 8.0 - 20.0

Clm Msw
3yr 3up 4up Male

Purse All
Vi 10 - 37
PL - 4

5.5 - 9.5
DirtFast Artif iDirt

K1 - K9

HTR = 1,2
RS = E, P, S, R

EPR = 080 - 099
TRN = 060 - 250
JKY = 060 - 250
PED = 350 - 990

FSize = 07 - 14

Not FIG2 = 1
Not BEAT FAV
Not bo

Thank you Ken, hope your weekend's going well.

[km]

Mark #5
+60%, 833 plays, 12% winners - good try!

Play All 00833 00101 12% 23% 34% 1.60 1.17 1.06 $26.5 0056 $122 1.11

01Feb06 - 31Jan07

PL-4
Clm Msw
3yr 3up 4up Male
Vi 10 - 37
Dist 5.5 - 9.5
DirtFast Artif iDirt
MLO 8 - 20
HTR = 1,2
RS E, P, S, R (No "F" or "N")
EPR 080 - 099
TRN 060 - 250
JKY 060 - 250
PED 350 - 990
FieldSize 07 - 14
NO FIG2 = 1
NO BEAT FAV
NO "bo"

[Mark]

Good morning Ken, you had a much higher play count than I'd expected, perhaps due to my convoluted presumption of "exclusions" deleted seperately as a Robot tester would.

Can the test be run as NO FIG2 = 1 or NO BEAT FAV or NO "bo".

Also, would you please rerun with NO FIG2 = 1 as the only exclusion, adding a NO CHAOS race filter.

My apologies for the confusion, especially if I still am.

[MikeDee]

I have been monitoring a selected group of the posted spot plays and yesterday they did quite well, 18% wins and a ROI that exceed 1.5.

The problem was that the two biggest wins were at Will Rodgers Downs and with out those two winners the group of spot plays was in the dumper.

Don't know where your wager, but WRD is not available to me. With spot plays even when you win you can still loose:(

[km]

Mark #5 Redux

Can the test be run as NO FIG2 = 1 or NO BEAT FAV or NO "bo".


That's how i ran your test originally Mark, any one of those three items would eliminate the horse, i doubt any horse would qualify for all 3 items at once! 1.60 ROI stands for that one.

I changed your test to NO CHAOS and NO FIG2=1 (and removed the other two NO items)

ROI = 1.18
Plays 800+
Win 11%

Much lower ROI, but also the test came in at 10.5%, so you barely made the minimum win%. Better leave the Chaos races in there as they produced the highest priced winners that made your ROI so strong (1.60) - so try again from the original Mark #5 as you have some plays to work with - but watch that win%.

[km]

Mark #5 additional comments and stand by for re-test

Mark asked me for a re run on his #5 play. When i ran his basic parameters through the Robot i got a better result, so i think i may have used the default PL-5 instead of the PL-4 for his tweak tests. I'll do it over and try to figure out if that was the reason.

BTW:
Contest ends at Midnight Friday EST

[km]

Found my mistake with Mark #5 with the first test i ran (1.60 ROI).
It included WET tracks, but the "wet" was supposed to be eliminated.
My mistake, sorry Mark.

So here is the first Mark #5 tested again =

ALL 00717 00096 13% 24% 34% 1.77 1.25 1.08 $26.4 0051 $122 1.23

Excellent play 77% return, 717 plays, 13% winners. Very close the leader (84%). A few more tweaks and it might get there Mark.

Robot users, you can largely re-create this one.

MLO 8.0 - 20.0

Clm Msw
3yr 3up 4up Male

Purse All
Vi 10 - 37
PL - 4

5.5 - 9.5
DirtFast Artif iDirt

K1 - K9

HTR = 1,2
RS = E, P, S, R

EPR = 080 - 099
TRN = 060 - 250
JKY = 060 - 250
PED = 350 - 990

FSize = 07 - 14

The only item that needs to be queried or programmed is
Not FIG2 = 1

[Mark]

Thanks Ken, 1.77 is a lot easier to work with.

May I request the same parameters with the following additions:

Race Filter: NO State Breds

Other Exclusions: NO JKY= 1,2 NO " bo"

[km]

You have done it Mark!!
91% return with 631 plays, 14% winners

You have the lead and the $500 unless someone comes up with a doozy.

ALL 00631 00088 14% 24% 35% 1.91 1.30 1.12 $27.4 0049 $122 1.29

MLO 8.0 - 20.0

Clm Msw
3yr 3up 4up Male

Purse All
Vi 10 - 37
PL - 4

5.5 - 9.5
DirtFast Artif iDirt

K1 - K9

HTR = 1,2
RS = E, P, S, R

EPR = 080 - 099
TRN = 060 - 250
JKY = 060 - 250
PED = 350 - 990

FSize = 07 - 14

NOT FIG2 = 1
NOT JKY=1 or 2
NOT bo

[DanG]

DanG #3

(Actually a modified #2 [800+ plays] after I re-exported all data with Ken’s 2007 update. Made a big positive difference in several ratings and I recommend it highly if you haven’t already.)

Paceline-5

Rating: Criteria: Description:
rPED : =1 Best parents in the field
nLONG : >0 $ and $$
nSEX : =1 Males
rMLO : >1 A spot play standard
nMLO : <>15 Impacts some horrific circuits
nFLD : >5 and <12 Not too small, not too big.
nDIST : >5.5f, and <>7 Not too short and 7f can be a bastard.
nLDIST : <11f The turn backs of 12f to 6f just didn’t work out?
nDS : <>5 No artificial sprints
nQP : >1 and <8 Eliminates extremes (A common theme)
nQT3 : <24 Eliminates wicked triple QP-8's
nQFIVES : <8 Eliminates extremes
nRS : <>4 Close up…or…way…way back.
nLFST : >5 Racing experience
nEPR : >97 and <111 Mid – strong class races
nK : >68.49 Scratches the bottom feeders
nHAG : <9 Sorry John Henry
nCHG : <4 Certain equip changes fail in many plays
nFPSX : <>3 Took out RS “S” and he dominate FR3 was poor
rHTR : <>5 Yes, NOT 5. I know it’s shocking.
nWGTCH : >-10 and <8 Eliminates extremes in LBS assignments
nLFIN : >0 and <10 Few “Bogeys”
nTJ : <=35 Sorry Russell B. & Jerry H.
nJKY : >90 and <415 Not the bottom and no Ramón Dominquez
nCLCH : >-15 and <15 Eliminates extremes
nPDIF : <11 Scratches the bottom feeders
rFR2 : >0 and <10 Needs ratable velocity line
rFR1 : >0 Also, needs ratable velocity line
nTOPPACAGO : >5 and <800 Needs best PAC # in this decade!
nTOPPERAGO : >5 and <800 Needs best PER # in this century!

BTW:

Looks like overkill?

This play won in 2005
Won every month of the last 12 (except Aug. it was $0.93)

Again...Still a work in progress getting things to line up! :D

[Rick]

Use Notepad to line it up and use spaces instead of tabs and spaces.

Can take some time but it does the trick.

[DanG]

Use Notepad to line it up and use spaces instead of tabs and spaces.

Can take some time but it does the trick.

Thanks Rick, but I’m not making any promises!!! :)

[njcurveball]

*** Congratulations Mark! I am once again looking at your hind quarters! ***:eek:

Ken,

Can you please take out the ML = 5 horses if you have not done it in the previous test.

Also can you take out NJKY < 75

Females
Mcl
E/P = 1
MLO 6-25 **** ML > 5 and ML < 25
MLO <> 10
RS: E, P, S
QP > 0
L/P > 3
LFIN > 3
QFIVES > 0
WK > 66
Field Size > 6
NJKY < 75 ***** take out the really bad jocks

thanks,
Jim

[km]

Dan's latest - Great ROI (+84%) but i just missed the 600 play count (just barely) - what did you get Dan?

ALL 00593 00108 18% 29% 40% 1.84 1.16 0.98 $20.2 0037 $114 1.44

Basic play is this =

PED=1
$ and $$
Males
MLO rank > 1
MLO<>15
FLDSIZE >5 and <12
DIST> 5.5f and <> 7

Dan's additional filters in his own words =

nLDIST : <11f The turn backs of 12f to 6f just didn’t work out?
nDS : <>5 No artificial sprints
nQP : >1 and <8 Eliminates extremes (A common theme)
nQT3 : <24 Eliminates wicked triple QP-8's
nQFIVES : <8 Eliminates extremes
nRS : <>4 Close up…or…way…way back.
nLFST : >5 Racing experience
nEPR : >97 and <111 Mid – strong class races
nK : >68.49 Scratches the bottom feeders
nHAG : <9 Sorry John Henry
nCHG : <4 Certain equip changes fail in many plays
nFPSX : <>3 Took out RS “S” and he dominate FR3 was poor
rHTR : <>5 Yes, NOT 5. I know it’s shocking.
nWGTCH : >-10 and <8 Eliminates extremes in LBS assignments
nLFIN : >0 and <10 Few “Bogeys”
nTJ : <=35 Sorry Russell B. & Jerry H.
nJKY : >90 and <415 Not the bottom and no Ramón Dominquez
nCLCH : >-15 and <15 Eliminates extremes
nPDIF : <11 Scratches the bottom feeders
rFR2 : >0 and <10 Needs ratable velocity line
rFR1 : >0 Also, needs ratable velocity line
nTOPPACAGO : >5 and <800 Needs best PAC # in this decade!
nTOPPERAGO : >5 and <800 Needs best PER # in this century

[km]

NJ's latest tweak, didn't quite make the 600 minimum plays, but the ROI just about did (+89%) with 16% winners.

Play All 00591 00092 16% 25% 36% 1.89 1.19 1.06 $24.3 0042 $140 1.49

Females
Mcl
E/P = 1
MLO 6-25
MLO <> 10
RS: E, P, S
QP > 0
L/P > 3
LFIN > 3
QFIVES > 0
WK > 66
Field Size > 6
JKY >= 75

[MikeDee]

In Mark5 what is the export equivalent to "NOT FIG2 = 1"?

Is it rPFIG <>1 1?

[DanG]

Dan's latest - this is the first spot play that took me more time to setup than it did for the computer to run it :rolleyes: Great ROI (+84%) but missed the 600 play count (just barely) with all those restrictions in tow.

Even after re-exporting everything its amazing how different our results are.

60 play difference and more than 30% in ROI? :rolleyes:

[km]

Mike - correct: nPFIG <> 1

Dan - no problem on the retest

I'll go through it line by line on the criteria and see if i missed anything. If a second test comes up the same, i'll make a list of the plays from February and we can figure out where we don't match.

Thanks for the comment on the $, $$ in the new export.

[Zaf]

Dan,

Love your spot play and your reasoning behind each factor. Great stuff !!!

[km]

Dan's Play Redux

I agree with Zaf - what an innovative set of eliminations Dan!

Found my error in running the previous test.
He had specified nCHG < 4, but i typoed it as nCHG > 4 when i ran it the first time.

He takes the lead = 96% return, 19% winners (!) and 641 plays - wow

ALL 00641 00119 19% 29% 39% 1.96 1.19 0.98 $21.1 0042 $114 1.47

PED=1
$ and $$
Males
MLO rank > 1
MLO<>15
FLDSIZE >5 and <12
DIST> 5.5f and <> 7

nLDIST : <11f The turn backs of 12f to 6f just didn’t work out?
nDS : <>5 No artificial sprints
nQP : >1 and <8 Eliminates extremes (A common theme)
nQT3 : <24 Eliminates wicked triple QP-8's
nQFIVES : <8 Eliminates extremes
nRS : <>4 Close up…or…way…way back.
nLFST : >5 Racing experience
nEPR : >97 and <111 Mid – strong class races
nK : >68.49 Scratches the bottom feeders
nHAG : <9 Sorry John Henry
nCHG : <4 Certain equip changes fail in many plays
nFPSX : <>3 Took out RS “S” and he dominate FR3 was poor
rHTR : <>5 Yes, NOT 5. I know it’s shocking.
nWGTCH : >-10 and <8 Eliminates extremes in LBS assignments
nLFIN : >0 and <10 Few “Bogeys”
nTJ : <=35 Sorry Russell B. & Jerry H.
nJKY : >90 and <415 Not the bottom and no Ramón Dominquez
nCLCH : >-15 and <15 Eliminates extremes
nPDIF : <11 Scratches the bottom feeders
rFR2 : >0 and <10 Needs ratable velocity line
rFR1 : >0 Also, needs ratable velocity line
nTOPPACAGO : >5 and <800 Needs best PAC # in this decade!
nTOPPERAGO : >5 and <800 Needs best PER # in this century

[njcurveball]

Congrats Dan!

What an incredible spot play! It is like watching Clemens with an umpire giving him the outside corner! Looks untouchable to me!!!

AND it is nice to know you did even better when Ken tossed that out those eliminations with "only" an 84% profit! Good thing he found them, they were dragging you down. :D

Thanks for sharing, I learned a lot just by reading what you had done!

Jim

[DanG]

Thanks Ken, Jim and Zaf…

Jim, you mean umps give the “rocket” a couple inches off the edge? I’m shocked!!! :eek:

In my personal play I never have a play for more than one track. I have one particular play and about 25 versions for each individual track. In addition I almost always use an Access query that is 5 or even 6 lines deep in criteria. On line for each running style.

For example…

To apply the exact same layoff criteria to a RS “R” and RS “F” is comparing apples to oranges. Not to mention mixing turf, dirt, artificial.

Try this little spot play (not quite enough plays for this exercise)… I played a modified version of this and frankly did very well at Hollywood this past meet.


Surf =4 (Artificial)
LSurf =2 (Turf to Artificial)
rHTR <4 (1, 2 or 3)


3 criteria and you get…20% W, 35% WP and $1.04 ROI (984 plays total)... (BTW: Only a $62 max)

That’s a nice ROI and / or a nice first three moves in chess. :)

[qhrick]

Dan, thanks for sharing your ideas
Common logic would prob dictate your currently leading play has far too many filters to be valid
Yet, an explantion is given for each, lending to validity

Another example of the value of this thread. Gets people thinking in new directions

Kudos on taking the lead

Just be prepared for the inevitable late entries, jumping in at the 1/8th pole like Carmouche coming out of the fog!!
-rick

[Mark]

You're a tough act to follow, Dan. I can add a few points to my ROI but probably not enough.

It's been a hell of a race regardless, we all benefit by the company we've kept.

Congratulations!

[DanG]

Common logic would prob dictate your currently leading play has far too many filters to be valid
Yet, an explantion is given for each, lending to validity

Just be prepared for the inevitable late entries, jumping in at the 1/8th pole like Carmouche coming out of the fog!!
-rick
LOL Rick, yes I know its FAR from over!!! :)
---------------------------------------
Thanks Mark and Rick…

Rick, you may very well be right, but as I said…This play won in 2005 and won every month of the last 12 (except for Aug…$0.93)

BTW: Last 10 days I only show 8 plays, 2 wins for $1.63

Would I play it?...As I said I make spot plays on a per track basis only. But, I will be curious to go back this time next year and check out several of all our efforts.

BTW-II…Here’s the track breakout concerning the (turf / to artificial and rHTR <4) play…It speaks to the point of being track specific.
Track Sample Win% ROI Net
HOL 146 25% $1.33 +$64.00
KEE 96 22% $1.30 +$52.00
TP 502 17% $0.84 (-$160.00)
WO 250 21% $1.18 +$89.00


So, in hindsight a $45.00 net profit was made betting every situation. TP reduced the overall net by 78% by being approximately half of the wagers and under-performing.

To me, this type of research is not limited to spot plays. These are the types of situations that apply to day to day handicapping. If you’re putting together a pick-4 at Hollywood and a strong turf finisher is now routing on the cushion you give them a 2nd and 3rd look because your research has shown they are outperforming the population on the whole.

Will the betting population catch up at Hollywood this year?...maybe. As Ken said when he first programmed FR1 (when the overwhelming majority of horseplayers said EP was FAR superior) he needed a wheelbarrow to bring home the cash. Horse players have dug into that at an almost perfect rate of .01 cent per year for the last 10. They don’t always catch on as fast as we think.

Arlington & all the California tracks will open up a whole new opportunity for us to stay ahead of the curve. That’s why it’s so good to have someone like Ken who specializes in looking ahead.

[Zaf]

Thanks for sharing your views on creating & researching spot plays. I have been working for a whole year on constructing spot plays and I have done reasonably well, but I have seemed to hit a wall as of late. I can't get to the next level. I feel after digesting your material my game should move up very nicely ( I hope ). Thanks for stimulating ideas and saving me an immense amount of time in this endeavor. It is very generous of you to share this with all of us. I would like to thank all others who have submitted their winning plays for all of us to see. I am proud to be a member of the HTR family.

[njcurveball]

I am whipping and driving and for those Harness fans, I am kicking the horse (a Jackie Mo trick). Let er rip Ken! And thanks for this contest, I use to do queries looking for a 20% profit, I would say my "muscles" have expanded greatly now!


*** RACE CONDITIONS

NSURF = 2 (TURF!!!!!)
NPURS > 9 (Ken's 10k cut-off)
NFLD > 7 (Larger fields, better prices)
NVI < 34 (This number does some great things!)

*** HORSE CONDITIONS (selection criteria)
NQP <> 1 (One quirin speed point was an eliminator in many tests)
RTRN < 4 (Top Trainers, somethings I always look at on turf!)
RCLASS < 6 (Need some class!)
NMLO > 5 ( Very hard to get an ROI over 50% without using this.)
RK > 3 AND RK < 9 (RK is a lil too good to eliminate some, but we need prices!)
RPER > 1 (looking for prices)
NPPO <> 1 (the one post is notoriously bad on turf)
NPPO <> 6 (This is a pure sniper condition)
NWKSC > 0 (Need to show a workout score)
NLDIST < 8.5 (This was an amazing elmination I was not expecting to be good at all!)
NPED > 80 AND NPED < 650 (I am starting to like this "taking out the extremes" theory)
NJKY > 140 (Bad jocks need not apply. The amazing thing was that the lowest ranks were profitable, but the ratings still made for a good eliminator of a few losers)
NWGTCH < 6 ( A small eliminator, but worthwhile)
NFPSX <> 2 (Large 2nd fractions do not help on the turf)
NFTI < 55 (Ken has a good number, seems it picks out the nicer breds who get bet)


Just for clarification, the "not equal" conditions

NQP not equal 1
NPPO not equal 1
NPPO not equal 6
NFPSX not equal 2

[DanG]

“Say hello to my little friend”!!!

http://www.starstore.com/acatalog/scareface_little_friend-01.jpg

Jersey Jim firing both barrels!!!

Knowing Jim it will be an $8.00 ROI and he will even throw in free Waffles and Ice Cream from the Seaside Boardwalk!
:D

Best of luck!!!

[MikeDee]

Here is my second and final attempt at the brass ring.

My spot play Mike2 is a access based play and is is attached.

It is uses the htr2 exports all_hx4 and all_hx7.

These are joined by the fields date, track, race, and program number.

The query is made up of 4 OR clauses. The key elements are based on the race conditions field found in all_hx7. The 4 key elements that the 4 OR clauses are built on are:
NW male,
NW female,
GRs and Stk
MSW

NW, Gr and STk are all lumped into nCla =1. So to split them for this spot play the text filter "Like" is used.

This is first and foremost a back fit to try and hit a ROI of over 2.0. No need to try and justify why certain filters are used except to say that they are there because they in some way improved the ROI.

Rather then to type all of the lengthy filters in this space I have attached a excel spread sheet. It is set up like the design grid of a query.

The actual filters for each of the htr exported fields are in columns B through AP

Row 2 of the spreadsheet is the first clause. Each column of B through AP are AND terms of each OR clause, (there are a couple of mini ORs thrown in Like Nclass 0 or >=77)

Row 3, 4, and 5 of the spreadsheet are the other OR lines of the query.

So one way to state this is (row2 col b thru ap) OR (row3 col b thru ap) OR (row4 col b thru ap) OR (row 5 col b thru ap).

Not every field applies to every OR clause. If the field does not apply in the OR clause it is blank.

Ken thought Dans' query was bad to code. He will be cussing me tonight.

If anyone wants to see want a 4 line query on joined tables looks like I can post the SQL version of this query for you.

[DanG]

Ken, as a “preemptive strike” :D can I tweak my last play #3 with the following conditions excluded.

• <> MCL (or Cla <>4)
• <> Purse <10 (Or all Purse >=10)
• <> “Bogeys” or <> “@” as it’s referred to in the Robot.

Stupid me, I didn’t know multiple line queries were kosher. Those are the type that I use everyday. Too late to start a multi-line one now, so I have to hope this last tweak push the last one.

Best of luck to all and thanks again Ken for a great challenge. :)

[njcurveball]

Looks like everyone is bringing out their big guns. So let me add another entry. I will keep this straight forward to save Ken a few hours work. :D

I have to admit that the more I worked with the FTI and VI numbers, the more impressed I became with them.

ncla = 4
AND NPURS > 5
AND NDIST > 5.5
AND NDIST < 9
and nsex = 1
and nsurf <> 3
and nvi > 26 and nvi < 41
and nchaos = 0
and nwksc > 65.99
and nFTI > 29 AND nFTI < 60
and nmlo > 5 and nmlo < 30
and nmlo <> 10
AND NLFIN <> 2
AND NLFIN <> 3
AND NLFIN <> 4
AND NWGTCH < 6
AND NTRN > 79
AND NQP < 5
AND NLFST <> 3
AND NLAY < 140


p.s. If you read this Mike I would like to see the code, maybe you can Email it to me or post it? Thanks.

p.s. (II) I think we proved Dan's tag-line to be correct with this contest. :D

[AwolAtHTR]

---
Ken thought Dans' query was bad to code. He will be cussing me tonight.



and almost guarantees that future Spot Contests will be Robot ONLY!!--LOL--

AND, reminder about the power of the Robot but limited sniper power.

so, if you do not need sniper power, then just use the Robot!!

aaah, good luck Mike, I hope your ROW breaks the two mark. Return-On-(Wager)

[Mark]

Two eleventh-hour submissions/ revisions.

Mark #6

MLO 8.0 - 20.0

Clm NO State-Bred
3yr 3up 4up Male

Purse ALL
Vi 10-37

PL-4

5.5-9.5
DirtFast Artif iDirt

K1-K9

HTR= 1 or 2
RS= E, P, S, R

EPR = 080-099
(K) =069-103
TRN = 060- 250
JKY = 080-250
PED = 350-990

NOT FIG2 = 1
NOT JKY = 1 or 2
NOT "bo"
NOT JT> 30
NOT Beat Fav



Mark #7

same as above with:

Fieldsize 07-14

Beat Fav OK (No Exclusion)

[Rick]

MikeDee,

Post the SQL view of your query.

Thanks

[DanG]

DanG # 4 and the grand finale. (All of the following is one play, broken down by angle / grouping)

Off to Gulfstream tomorrow and best of luck to all with an excellent card.

HTR Staples:
Rating: Range: Description:
rPED =1 • Edge in the ‘gene pool
nPED >=320 • Worth some money
rHTR <5 • Automatically includes at least 70% of the winners
nK >65.49 • *Excludes the poor souls who should pull wagons
nLONG =2, $$ • Ken’s tweak to this longshot rating is powerful!!!
rVEL <10 • Ensures at least a little gas in the tank.
nPDIF <13 • * See the wagon pullers.
rIMP1 <10 • Within distance limitations I LOVE the impact ratings
rIMP3 <9 • Need the horse to demonstrate a little “resistance”
rIMP4 <10 • Includes a more balanced attack

People:
Rating: Range: Description:
nTRN >=80 and <=425 • Excludes incompetence and the ones with a siren on the bridle.
nJKY >=125 and <=355 • Excludes the ones who can’t stay on and Russell “the muscle”

Form:
Rating: Range: Description:
nLWINF <1 • The public will always love a good last race.
nWMAR <=1 • The public will always love a horse that wins by a large margin.
nLFIN <12 • Trying to reduce “bogeys” where they at least finished,
nLPOS1 <12 • Trying to reduce “bogeys” where they at least tried early.
nLFST <>1 • No 2TS

Race Composition:
Rating: Range: Description:
nCLA 2 or like 4 • Ran for a tag, Claiming or MCL
nFLD >6 and <12 • Impacts a few No Cal races and a few wild CNL and LRL MCL.
nVI >21 and <40 • Trying to avoid the inscrutable and Bernardini in a walk over.
nQT3 >6 and <24 • This and Q5 are very underrated ratings
nFTS <4 • Trying to reduce Chaos
nRUNKS <6 • Trying to reduce Chaos…again.
nCHAOS =0 • REALLY trying to reduce Chaos
nTAN <3 • Love when horses meet for the first time, confuses the public.

Distance: (I know the robot allows certain dist structures. Not sure if this is written correctly, but here it goes.
Rating: Range: Description:
<>6.5 • Those 4 excluded and “NO” <3.5f to exclude the early MCL and bush sprints.
<>7
<>7.5
<>9

Recentcy:
Rating: Range: Description:
nLAY <600 • Within the millennium
nLAY90 <>2 • 2nd off layoff does surprising poorly in MANY circumstances.

Trk Specific: we can’t be track specifics per-say, but these variables address a circuit somewhat.
Rating: Range: Description:
nCLM >3 • Cuts out many minor tracks who run for as low as 1k
nPURS <44 • Claming stakes or maybe inflated state bred pots.
nSURF <>2 • Bye Bye to my favorite surface.
nCOURSE <>3 • Bye Bye to the Inner track.

Horse:
Rating: Range: Description:
nLFWP <35 • Too many bells and whistles go off with these “win types”
nLWINS <8 • Too many bells and whistles go off with these “win types”
nHAG <8 • I was always taught to respect my elders, but not always with horses.
nCHG <>2 • Blinkers On rears its statistically ugly head again!

Position:
Rating: Range: Description:
nPP <11 • Gate is not positioned in the parking lot.
nQPC <46 • Does not have a huge obvious QP advantage.
nGATEBRK <>1 • Did NOT break like a jack-rabbit in their last alerting the public.

Details:
Rating: Range: Description:
rMLO >1 • Self explanatory and in most spot plays
nClassFit <2 • No Class fit minus or “3” which is not defined yet, but with upside.
nWGTCH <10 • Jockey did not over-indulge during Thanksgiving
nFPSX <>2 • Not the dominate FR2 # which can be horses who lose contact early
nCLCH >= (-10) • These are horses for a tag, trying to avoid the bad” For Sale Signs”.
nTOPSHFAGO <535 • Let’s give them a year and a half to run their best number.
nFC <>65 • So often a “dead zone”
rCLASS <9 • Much rather be ranked “0” than at the bottom.
nWKSC <=88 • All running for a tag, maybe it’s not best to explode in the AM?


In closing a major thank you to Ken for staging this exercise. Very informative and thanks to all who participated and offered their thoughts. :)

[njcurveball]

WOW! That is the WHOLE enchilada Dan!!

Now I feel lazy with my effort! :(

Well at least with 3rd place I will get my entry fee back. :D

Great job fellas! The ideas you have shared are worth much more than the prize!!!

Jim

[km]

The Contest is Closed and so is this incredible record setting thread!

Thanks to all of you for participating and for all the hard work.
Mostly we are grateful to all that shared their insights.

It is going to take awhile to evaluate all the plays submitted in the last 24 hours.

The $500 prize (almost seems anti climactic now) will be awarded after we test and retest and verify all the final entrys.

The contest results and comments will be posted in a new thread "Spot Play Contest Final"