Blog Dec 16

[km]

I'll share some more features of Robot II as i work on them. And DanG gave me a great idea with that TAM shipper listing he posted.....thanks Dan

When you run the Robot' LEARN MORE on a single track, the Tracks listing is useless because only one track is being tested anyway. Sooooo - why not re-program that to "shipping" and then record the all data as shipping from every track. This could be quite interesting and since testing a single track in the Robot is the fastest mode, it would be something we could all grab in a hurry too. Kudos to Dan for another good idea :)

In another thread, storreyrich, asks about the various single-use Race Filters in the current Robot. These include Chaos, State-bred, Rule of 50, etc. These items are all checkboxed in the new RobotII and you can include any or all of them at the same time or you can choose the opposite and remove them (No Chaos, No Statebred, etc).

Robot II expands the rankings to show 1-2 ranks for all factors including every one of the Impact ratings. And goes all the way with 1-9 ranks with the HTR Consensus rating (HTR = 1-9) just as with the K ranks. Also, the full spectrum of Quirin Points is available.

I have added some key new filters and items that should revitalize test results with pace and speed numbers - below is one idea.

RobotII adds a number of checkboxes that instantly remove negative items from your test or spot play.

NO Chronic Losers (my favorite because i hate getting a spot play and finding out the horse is 1 for 39)
NO Bad Trainers
NO Bad Jockeys
NO Bad Pedigrees
NO Bad T+J
NO Cold Trainers
NO Bad Fr3

and more
These can have a tremendous impact on improving spot plays or angles by eliminating the weak links in the data chain. They can be chosen in multiples or you can check-off all of them at once if you want. That would be a very complicated programming chore for a db user to configure and test all of the permutations of these, but is a snap in Robot II with the click of the mouse.

[Bob]

Ken,

Fantastic changes...thanks so much!

Question:

Would it be difficult to have buttons for a rating like say, Average Pace, front half of field?

[km]

Sure, you can use RS or QP assignments to get a good tandem with "front half of field" Bob, they are right there next to the A/P in the new Robot. You can experiment easily with that now.

[Bob]

Ken,

Fantastic!

[edw]

RobotII adds a number of checkboxes that instantly remove negative items from your test or spot play.

NO Chronic Losers (my favorite because i hate getting a spot play and finding out the horse is 1 for 39)
NO Bad Trainers
NO Bad Jockeys
NO Bad Pedigrees
NO Bad T+J
NO Cold Trainers
NO Bad Fr3

Ken, if the information is proprietary and following is inappropriate to ask, I'll apologize in advance but would it be possible to let us know the exact criteria RobotII is/will be using to define the above? If necessary, perhaps post it in the member only area? The newsletters and past posts provide a lot of insight into what those criteria are likely to be but I'd prefer to know with certainty if possible.

Thank you.

[km]

Yes, that's the plan edw, i'll put the criteria in the newsletter or user guide available in the Subscriber Zone

Example Bad Trainer: TRN < 150

[edw]

Thanks, Ken. We all appreciate how you regularly make our handicapping lives better and better. Maybe we shouldn't be so selfish and suggest to P-E Obama that you take charge of fixing the financial crisis! :D

[Gar-Bear]

KM-would it also be possible to add to the negative items "No @ horse". I have found from my testing that most these horses are bad and would improve the results of many of my spot plays if I could eliminate them. Thanks for all that you do. The Robot II sounds great from the little bit that I have been reading.

[storreyrich]

so many great ideas; like the chronic losers filter ; I have a fairly lenient rule of thumb that 10 + race maidens can never be bet and horses who can't win at least 10 % of their races likewise. Rarely misses any worthwhile prices because these horses surprisingly are usually underlays because they often have competitive Beyers and decent ITM %'s ; they just dont want to win; I ascribe that to the pack mentality of horses; that's for another day but these horses just feel comfortable losing; they often get the lead in the stretch but wait for someone else to come and take the lead because it just makes them feel better.

[DanG]

I know your is well into this project, but just throwing things out there without knowing if any of its possible.

• Turn back
• Stretch out
• Trainer Change / positive or negative.

Actually all of the above can be linked to the trainer download for R to S, TC stats assuming the sample is OK. Also; on the turn back…7f to 6f can be a nice angle and doesn’t get covered with the more traditional “route to sprint”.

• Paceline-4 indicator: Just an alert flag that the animal has a rating with PL4 / or not.
• TPG (If the + - makes too many variables, just the grades)
• Any type of PPX information. (No idea how to program that one :confused: )
• “Power Move” – a sudden burst of 3+ positions pre-stretch call.
• Any factor the “eye-ball” highlights.
• Paceline-5 alert: If ‘PL5 skipped over a good dist / surf / finish match because of a missing pace call. (I.e.….@ about distances, fog , other acts of God or Magna)
• Significant Claiming price movement off positive or negative effort.
• A custom Gupta fire alarm when he’s all in on a router at GG. :D

Thanks for all your brainstorming Ken; it is soooo appreciated and I hope you take time to enjoy your holidays with your family :)

[km]

Gar-Bear = that one is in there as "No Big Losers" any horse that lost it's last by 20 or more (or pulled up, vanned off, distanced, etc.)

storry - yes, i used 0 for > 10 for chronic loser maidens, < 6% lifetime wins for horses that have won a race or two (ie 1 for 20). These are the statistically proven numbers. But i tend to agree with you and toss almost any maiden who has lost 8 or more, unless they are dropping in class.

Dan - thanks for the ideas, i'll certainly consider them, excellent angles to include. The Big Class Drop is already included. Gupta (OPM) alert is under serious consideration!

i'll address two of your suggestions below that you won't see right away in the Robot II . TPG and PPX (compare last rating to today's rating).

TPG
The problem here is not the programming difficulty, but the overload on the processing and time consumption. TPG requires a lookup of the Trainer Text Stats file (same for Trainer Report Card grades). That file is not separated by single races and scratches are not removed - it must be parsed raw with a horse name lookup for the entire card. That type of text i/o won't take much time when looking at one race and computing the TPG, but it would slooooow the Robot to a crawl if forced to do it to thousands of races.

PPX
Even slower (much much slower) would be the PPX items, for example, comparing a horse's last race PED with today's race, especially if it is surface switch. Great handicapping angle, but the time it takes to lookup the previous (assuming the user even has the previous racefile in the current folder) would be super slow. You can even see this when clicking the PPX button, some races take 5 sec or more to load. '

In order to get at those PPX items, we would need to pre-process the data into a statistics binary file for faster lookup by the Robot. The TPG and Trainer Report Card grades are more likely to get into the new one - i guess we could be warned that the processing time will be slower before using those filters, i'll work on it.

[DanG]

Understood Ken;

Please don’t slow the processing time for a few bells and whistles. Just throwing ideas out there and as usual; when we see the finished product there will be things we never would have dreamed of.

HTR continues to be one of the few sources of continuous good news. :)