XFAV using HTR-1 vs. K-1

[nomadpat]

I was playing around with the robot using the XFAV parameters from the Dec newsletter. Substituted HTR-1 instead of K-1 and the results:

K1: 49% Win, .89 ROI, 72% WP, .94 Place ROI
HTR 1: 50% Win, .94 ROI, 74% WP, .96 Place ROI.

I don't have a large database since I'm relatively new to HTR, so I'm curious if someone else can run over a larger sample and report?

Thanks! :)

[km]

Excellent results Pat. What date range was your sample? How many plays were there in each of those samples (MY PLAY)?

HTR and K are obviously very similar.
In the long run the (K) is a more consistent overall production engine for WP%.
The (HTR) gets more longshots but with a lower win rate and some ties. (K has zero ties).
With XFAV - the pattern will be cyclical and favor both from time to time.

The main reason the HTR pulls more prices is that it weights the Wk and PED equally with the other data in the consensus. So if those factors are hot with XFAV; the results will be great with the HTR rating. You could also try the Wk and PED filters with the (K) and XFAV and you might get similar results. You can see some of my results with that in the newsletter.

[hurrikane]

The late great George Bloomfield also had a version of xFAV involving a pecentage of the top 3 K (old dos K) and Pscan PL1 and 5.I think his hit better than 50%. One of the few guys I've met that could make money betting favs

[nomadpat]

Thanks for the reply KM. My plays was 304 for K and 333 for HTR.
I have found that when I'm handicapping a race and I like a horse not in the top 2 or 3 K, he shows up in HTR and vice versa. So from these results, I will leave both as Custom plays on the Robot. Is there any chance we could name our angles in a future edition - instead of "custom 1, 2, 3, 4"?
K1 had a real doozy today at BEU. Race 3, the 9 was 9/2 MLO, 5/2 KLine and won at 16-1!! :eek:

[Donnie]

Don't know if anyone is struggling with this or not, but here is the sql for the query to find the XFav's using Access. Copy the code below, start a new query based on the HX4 export, go to VIEW to SQL VIEW. Paste in this code and run the query. This is with the extra filters found on Page 7 of this month's newsletter. Where the winner is found 40%, win/place 70% and ITM 80% of the time. Let me know if you have any trouble with it!


SELECT HX4.tTRK, HX4.nRACE, HX4.tPGM, HX4.tHOR, HX4.nMLO, HX4.nLAY, HX4.rHTR, HX4.nJKY, HX4.rLCSF, HX4.nCLA, HX4.rKrat, HX4.rSP, HX4.xPL, HX4.nPURS, HX4.tDATE
FROM HX4
WHERE (((HX4.nMLO)<=2) AND ((HX4.nLAY)<=45) AND ((HX4.rHTR)=1) AND ((HX4.nCLA)=1 Or (HX4.nCLA)=3) AND ((HX4.rKrat)=1) AND ((HX4.rSP)=1) AND ((HX4.xPL)=5) AND ((HX4.nPURS)>10)) OR (((HX4.nMLO)<=2) AND ((HX4.nLAY)<=45) AND ((HX4.nJKY)>=350) AND ((HX4.nCLA)=1 Or (HX4.nCLA)=3) AND ((HX4.rKrat)=1) AND ((HX4.rSP)=1) AND ((HX4.xPL)=5) AND ((HX4.nPURS)>10)) OR (((HX4.nMLO)<=2) AND ((HX4.nLAY)<=45) AND ((HX4.rLCSF)=1) AND ((HX4.nCLA)=1 Or (HX4.nCLA)=3) AND ((HX4.rKrat)=1) AND ((HX4.rSP)=1) AND ((HX4.xPL)=5) AND ((HX4.nPURS)>10))
ORDER BY HX4.tTRK, HX4.nRACE, HX4.tPGM;

[Rick]

The only problem with the query for most users is that they are using the database where the table name is ALL_HX4 instead of HX4.

Anyone going to use the query and has table named ALL_HX4 could probably do a search and replace in WordPad. Or you could rename them manually.

[CC Brown]

Thanks to Donnie and Rick. I copy the SQL to wordpad did a replace and it worked great.
49% win rate .92 roi.

[Donnie]

Hey CC...how 'bout you post that out here for others? It's Sunday...I'm too lazy!!

Now the key is to find the ones that will win and why!!

[CC Brown]

Actually for now all I want to know is which races these are and avoid playing them because I am looking for overlays where the ML favorite is vulnurable. If the ML fav is an XFAV I will just pass that race and try another one.

So now I run my picks for today then look at the XFAV for today and cut the XFAV races.