Contender Selection?

[Dutch]

Does anyone have a quick-and-dirty contender selection method?
Some of the most success I had in the past, was using a computer program that had manual input (don't remember the name, but I got it free as a beta tester, thru PRN - and then the developer died). Since it was manual input, I didn't want to HC all the horses, so I tried various methods to narrow the contenders. Then I used the program on just those contenders, and either used the odds-line generated for overlays (my favorite) or played the top pick. As I remember, I got some nice overlays without too much work. But you have to have a good contender selection method, or else you miss some good horses.
I used to have a very involved method using ITM and trainer stats and speed factors - but that was worse than computer input.
Oh, I forgot to mention, sometimes you needed to use more than one paceline - another reason to narrow your contender search. With manual entry, multiple pacelines add to the work.

[andicap]

Try this one on for size:

1. Go back 90 days.
2. Note third best Beyer. That's your "race par." Any horse with a figure within 2-3 points of that is a contender.

3. Consider all horses last three races at today's distance structure/surface (dirt-dirt,) fast track only (although you could use "good" or "Wet fast" in a pinch. On the turf, good is okay.)

4. do not go back more than 180 days.

5. If horse has more than 3 races that fit within 90 days, go back to a 4th line if its within 90 days.

6. FOr horses off layoff, or 2nd or 3rd off a layoff, be more lenient. These are judgement calls because it depends on various factors:

-- I don't like horses off of 180 days layoff or more.
--- Horses that run excellent races off layoffs (acc. to Beyers) often regress in their next race. Wait until they seem to be cycling back to their best figures.
-- Horses that show poor races off a layoff I ignore until they show SOMETHING that indicates they are rounding into form.
-- Horses that show some 'oomph" off a layoff are worth following. Could be early speed and fades, or a fast middle quarter or huge move in the end. Watch for "karma moves" after that first race that indicate the trainer is going for a win. (change of jockey, back to preferred distance, drop in class, etc.)

Best thing to do with horses off layoffs is know the trainer's habits. For example, everyone knows Dutrow is a killer off a layoff. Also see how horse has done in past off such layoffs.

7. Contenders: Take half the field plus one up to 5. Alternatively, you can only take top 3 or so if there's a huge gap between them and others. Be careful: often the "fifth horse" is the one that will pop at big odds.


Less effective on turf than dirt because figures tend to be much more competitive and horses don't "bounce" quite as much.

On turf conventional wisdom is play best closers (i.e. do the above but for best 2nd + 3rd call figures). Often true, but

a) check track profile: speed will sometimes hold up on baked, hard turf. Or in two-turn races, inside post horses will control pace and save ground. (trips play much bigger role in turf races especially around two turns. Certain jockeys can't ride in turf races because they require patience, good sense of pace, nerve and skill in shooting through tight holes. And although I've never tracked I would bet some jockeys have decent record in 1-turn turf routes, but poor record around two turns where you really need to save ground around the first turn.)

b) check the pace: speed often holds up when one horse has dominant early speed figures or in paceless or lone speed fields or sometimes on soft ground.

above for turf routes only. Turf sprints confuse the heck out of me.

I've sometimes suspected the best way to play turf races are to be patient and look for horses that were expected to contend but went 3-wide around the first turn and finished poorly, and are now back closer to the rail. Someday I'll test this theory.

[linrom1]

Never fail method is to use the horse that has the highest percentage ITM on turf.

[Tom]

Here's a Beyer contender selection given by William Scott (How will Your Horse Run Today) to a Sartin Seminar in Baltiomore around 1994:

1. Look at each horse's last two Beyers and circle the highest for each horse.
2. The top three numbers are contenders.
3. Ignore the last two Beyers for each horse and look at the rest of the races-circle the best Beyer of theses races for each horse.
4. The top three numbers are contenders.
5. You now have from 3 - 6 contenders.

[BIG RED]

Sounds like something my dad used for many a year, only he looked at how much money they made on the turf. Almost the same idea.But a good older horse would have much more money than a good younger horse, but then again that automatically gives you the more experienced horse. Maybe he had something there.

[Dutch]

Thanks, some good ideas. Funny you should mention Larry Scott, Tom. I recently moved and for the first time looked at and boxed up all my old handicapping materials. The one book I can see from where I'm sitting, is Investing at the Racetrack by Scott.
I like that Beyer method (or you could use it with BRIS numbers) - I might look at that.
Andicap, some good ideas there - maybe a little bit "overthought" with the other conditions. Are you a contributor to Jon Worth's Tips Report? Your name looks familar.

[andicap]

No one would ever accuse me of overthinking

:D

Nope, must be another Andy.

Yeah, you're right. I do have a tendency to go into all sorts of tangents in trying to get that win rate up a few points and not miss the losers.
That's a good question: would you sacrifice a few points for much more simplicity. Of course my method could even do worse than the ones with fewer rules!

[Shacopate]

Dutch,

I use something similiar (for a quick and dirty) that Tom and Andy have suggested. Both approaches are fine, so I won't offer mine here. (If you want it just send me a pm.)

What I suggest is that you do the same thing with pace figs.

Craig offers them for free (major tracks) and they are very good.

The two contender selection methods combined (pace and speed) will uncover some pretty good overlays if you make an odds line.