Track Profiles

[JimL]

I have recently reviewd my track profiles for the past year and I have come to the conclusion that all they really reflect is the universal track bias. What I mean by this is if a race shapes up with alot of early pressure then a P or S horse can win and it is best not to ignore closers just because a track profile says a horse must be within 2 lengths at the second call. Are track profiles an exercise in futility?

[Rick Ransom]

JimL,

My conclusion is yes, except in extreme circumstances. I used to put a lot of effort into trying to determine track bias but it didn't add anything to my bottom line and was a real pain in the butt to keep updated. I was extremely happy to find this because I was getting really burned out by never being able to take any time off.

What I found was that if you looked at short periods of time you could actually be hurt by keeping this information. Other players seem to be aware of the short term bias and prices go down. Also, at most tracks, if an extreme bias appears then the track maintenance crew will do its best to eliminate it. If you keep a profile for the entire meet, it won't hurt you but it won't help you either. You'll win a few more but the prices will be lower, so your ROI will wind up about the same.

I know will disagree with me on this, but that is how it works with my method. Pace was once a very important part of my method, but I have found that it duplicates other factors but is less accurate.

[JimL]

Rick, I want to thank you for your reply. I remember watching the replays from the New York tracks and Harvey Pack and some other astute NY handicappers would say,"What a speed favoring track today" Not true; the races usually only had 1 or 2 E horses! Bias I really question it.

[ridersup]

Call it what you want but certain conditions exist during the course of a tracks meet that you should pay attention to . You can call it track bias, configuration or maintenance. For example at the last Tampa Bay meet during the first 51 6f races there was only 1 winner from the 1 hole. There were many favorites breaking from this position that you could bet against and gain a significant advantage. Conversely last year there was a period when the 1 hole was deadly at 7 fur to the point that you could make money just betting this position.
Also when we have a big northern wind blowing into the horses face as they take that long run into the wind from the 7 fur chute you had better have your money on the closer. When the wind is behind these same horses and they get blown down the long back stretch the front runners usually prevail.

Warning. Don't bet too much money on an outside post position at CRC at the 1mi distance. Starting gate is almost on the turn and outside guys get carried wide.

During the week they put the rails up on the turf course at Churchill Downs and the feeling of most of the insiders there is that this gives the advantage to early or tactical speed. On the weekends with the rail normal the closers have a shot.

I read a study on Dave Schwartz's board that indicated that good results could be obtained by betting the 1 post at 1 mi and 1/16 at Pimlico. Dave you can correct me if Im wrong but I do believe it was on your BB.

In any event I believe there is significant advantage to knowing the track you are playing and using every tool available to you including trips, track variants, pedigree info, etc.



Ridersup

[andicap]

I agree with Ridersup. The biases to be aware of are more along the lines of how the rail and certain post positions are doing at particular tracks due to issues of drainage, maintaince, weather, track configuration, etc.
Nothing in stone, just probabilities, but anyone who has bet Belmont in the fall knows that a wide move -- the "Belmont Balcony" -- is usually necessary to win on the dirt.
I've also seen changes in turf courses due to the weather -- in periods of droughts, the courses may get hard and give advantages to front-runners. When it's been wet, the biases also change.

[Dave Schwartz]

Ridersup,

Yep. I posted that. And in about a week our query tools will be back completely and I can give you an update on that study.

Let's find out how many people read that article, huh?

Regards,
Dave Schwartz

[hurrikane]

I believe you gain a signigicant advantage with modeling. Of course there are lots of ways to beat the game. Unfortunately I personally don't believe you will find the ways if you don't at some time do the modeling.

I suppose it depends greatly on what info in you are modeling but if you don't understand what is going on at your track it seems to me you would be at a great disadvantage.

As for Pimlico..that has been going I would guess since Pimlico was built. Beyer and Davidowitz both talked about it in thier first boods. The run to the first turn in routes is short and Pimlico has very tight turns. This is a disadvantage to the outside horses. Great advantage to the 1 2 3 slot. Problem is..every one knows it so the price is reflective. Frankly I was surprised Point Given came off at the odds he did n the preakness with the outside post. After seeing that race he was a lock for the Belmont.
Just a note..that inside post doesnt' hold up in the rain...so my model says.

[karlskorner]

Is there a track bias ?

I quote from the track maintenance report for CRC for Monday and Tuesday.

Monday: What did you do to the main track today ?
We got another 1 1/2 inces of rain at Calder yesterday. We graded the track this morning to pull the dirt out that washed down inside and leveled it out. The floats and pack harrows were used aftrer the track closed to dry it out as much as possible..

Tuesday: What did you do to the main track today ?
We will use the packing harrows today. We got 1/4 inch of rain last night so we will not have to water the track early. We will water later in the day if needed.

What I am trying to bring out, the track changes every day at the whim of the maintenance crew. The bias that existed yesterday is gone today. The only bias you can depend on is to watch the first few races to determine if there is a bias.

The jocks try to read the track during morning "workouts", but than come the tractors and harrows and everything changes. Even between races the harrows can change the track. move the surface one way or the other. make the track deep or hard.

With this thought I question, bias, track profiles, post positions etc.

Karl

[Tom]

I think when you have data that shows the first two horses at the first call win 85% of the races at a certain distance, you can pretty much play it as if it was a bias.
But you still have that 15% that come from somewhere else and they might be paying good prices. That's where
the race shape bias comes into play. Right now, at Finger Lakes, there seemd s to be no real bias-horse all alone on the lead are quitting, horses involved in duels are holding on to win after fighting off the challenge and horse can come from downtwon and win. But there are days when the first one out of the gate wins. Period.
Trouble is, by the time you realize this, its over. I think real bias doesn't happen every day and doesn't last too
long when it does. I think the best use of bias is after the fact in analyzing what happened.
Tom

[Rick Ransom]

I'm not saying that biases don't exist, it just seem to be very hard to profit from them. The "universal track bias" seems to me to be that most horses win from just off the pace, that is E/P types. But horses don't run the same every race and it's way different to look at how the winners ran their races than it is to predict it in advance.

Some of the biggest prices on winners are on horses that unexpectedly run closer to the lead. Their past performances may have offered no clue that they would be that close. Maybe the trainer or jockey knows, but you don't. On the other hand, the E types with 1's all the way down at the first call will be bet down heavily. The public is aware of pace in an approximate sort of way.

As to post position biases, these would mostly affect distances where the starting gate is very close to the turn. Otherwise, there is plenty of room for the jockey to maneuver to get to where he wants to be. There is no requirement for the jockey to run the horse in a particular lane.

A well maintained track will not have a rail bias for very long. All of the jockeys will be well aware of it while it exists and run their horses wherever the track is fastest.
If the rail is slow, it will be really noticeable to the fans as well, because the frontrunners will be running out from the rail a bit. But they won't be able to take advantage of it because the jockeys on the inside horses will immediately take them out from the rail. If you want big time post postion biases, try harness racing.

I think though, that if you combine running styles with post position you might have something. Take for example two front runners in the 1 and 2 posts. If the 2 is a little faster out of the gate, he can get in front of the 1 and ruin his chances. You can think of all kinds of scenarios and it may not always work as anticipated, but the start is probably more predictable than any other part of the race.

[Dave Schwartz]

Ridersup,

Re-ran that query using just races from this year...

Get this:

LRL 1 1/16 miles
PP=1
Win Bets

47-180 26% $2.68

That's a 34% advantage!
Avg Pay=$10.26


Amazing, ain't it?

Regards,
Dave Schwartz

[hurrikane]

Careful. Dave is correct.
But before we all run out to Laurel and start driving the odds down on the 1st post...

there is an $80 bomb in there accounting for much of the profit. With my luck I would have been in the john when they closed the tote.

Good news is ..70% of all routes at LRL are won from inside the 4 post....80% from inside the 5.
In effect you could eliminte half of the field before they enter the gate.

Now...if you can make a profit playing short fields and chalk at the MD tracks...go for it. I live here and gave up on them a long time ago. An occational spot play or trainer pops up with a price..otherwise...you should really look elsewhere.

[Larry Hamilton]

Dave, you posted a 47 wins from 180 attempts for a resulting 26% win rate.

What if, in the 180 attempts you could find 100 favorites. What if in that 100 favorites you could safely identify which one of the favorites were likely to win and which were not?

My back-of-the-envelope guess would be you could crank up your win rate to ??%.

This is really neat stuff. You start with one problem--bias. Solve a little piece of it. Introduce a new problem--poor choice favorites. Chase it. Map and trace one problem solution onto another. Result could be we all move to maryland and buy Horse street....hehehehe

[Dave Schwartz]

Larry,

You could be right about the win pct but as Hurrikane mentioned there was (apparently) a big bomb in there. I don't have enough of the query tool working to look at the individiual payoffs yet, but I am sure he is right.

What amazes me is the hit rate.

Even in the original study (over 1500 races) the win rate was 23%.

For those that might be interested he is the original article.

http://www.horsestreet.com/freestuff/statistics/lrl85.html

Regards,
Dave Schwartz

[Rick Ransom]

Dave,

Is there a logical reason for the bias at this distance at Laurel? I'm skeptical when I can't explain things like that and tend to think that things will swing the other way in the future.

[Tim]

When I am running queries that are going to display both win percentage and average payout I have found it advantageous to eliminate 10% of the races. The 5% with the highest odds and the 5% with the lowest odds. If your sample size is not large enough to absorb this then the query results are useless anyway. What this does is eliminate the impact the bombs will have on the average win price and eliminate the impact of odds-on horses on the win percentage.
On queries with over 200 rows I increase the eliminated data to 20%, the top and bottom 10%.

Tim

[Rick Ransom]

It just occurred to me that track bias is like the fine print in those retail promotional offers you get, "subject to change without notice".


Tim,

Sounds like a good way to filter the data. I've heard of others using median price before, but your way is easier to do and, I think, more valid since it is something a player might reasonably do to avoid both low-priced underlays and hopeless longshots.


By the way, this is a subject that is not discussed nearly enough. What is the most accurate way to determine from a small sample whether you might have something that works? I've used win ROI for the top 3 selections, show ROI for the top selection, and many other methods including some that are pretty complex involving how much better I can predict finish postion than just by odds alone. What are your favorite methods?

[Que]

I'm showing the following for races at Laurel @ 1 1/16m on the dirt (last three years):

Post Position Results (Win ROI) by Odds & Higher (http://www.100megsfree.com/que/Picture1.html)

Post Position One (http://www.100megsfree.com/que/Picture3.html)

Post Position Two (http://www.100megsfree.com/que/Picture4.html)

Post Position One -- Post Time Odds & Higher (http://www.100megsfree.com/que/Picture6.html)

Post Position Two -- Post Time Odds & Higher (http://www.100megsfree.com/que/Picture5.html)

If I played Laurel, which I don't, I'd prefer to bet horses in Post Position 2 with post time odds between 10-1 and 50-1. Also, you can see that horses in Post 1 are heavily bet, but still produce a nice ROI in the 2-1 to 3-1 range.

Que.

[Que]

In addition to the post position study above, the chart below shows the average finish, and ROI to win, grouped by the horses running path for races ran at Laurel at 1 1/16m on the dirt. The data below verifies that horse's running more than two wide at the first turn have a very poor overall ROI.


Laurel 1 1/16m -- Dirt
AvgFin Path Count ROI

3.98 1w1w 871 0.785
3.93 1w2w 494 0.960
3.31 1w3w 301 1.265
3.71 1w4w 131 1.349
4.36 1w5w 49 0.783

4.18 2w1w 753 1.056
4.34 2w2w 3174 0.731
3.60 2w3w 929 0.864
3.86 2w4w 320 1.067
4.32 2w5w 64 0.421

4.79 3w1w 284 0.606
4.18 3w2w 544 0.669
4.23 3w3w 1196 0.747
4.16 3w4w 439 0.631
4.66 3w5w 98 0.316
5.08 3w6w 12 0.325

4.89 4w1w 118 0.500
4.48 4w2w 233 0.762
4.66 4w3w 447 0.722
5.04 4w4w 335 0.456
5.21 4w5w 82 0.668

5.18 5w1w 38 0.247
5.46 5w2w 78 0.411
4.75 5w3w 119 0.960
5.03 5w4w 96 0.089
4.85 5w5w 34 1.738

5.07 6w2w 13 0.361
4.71 6w3w 21 0.457
5.13 6w4w 23 0.230
6.38 6w5w 13 0.330


Looking at the above stats, the key to playing this distance is to bet the horse, or the jockey that can his mount to the rail by the first turn--otherwise, you are just tossing your money to the wind.

Que.

[Que]

Finally (am I bored tonight or what), the following table is grouped by post postion and running path, with the horse's average finish and ROI to win:

AvgFin Post Path Count ROI
3.85 1 1w1w 390 0.856
3.58 1 1w2w 186 0.913
3.07 1 1w3w 128 1.320
3.05 1 1w4w 51 2.101
3.65 1 1w5w 20 0.820
3.62 1 2w1w 145 1.140
3.56 1 2w2w 406 1.003
3.13 1 2w3w 81 1.280
2.79 1 2w4w 29 0.693
3.85 1 3w2w 14 0.899
2.78 1 3w3w 28 1.182
3.20 1 3w4w 20 1.400

3.91 2 1w1w 195 0.853
3.93 2 1w2w 132 1.341
3.37 2 1w3w 75 1.285
3.93 2 1w4w 32 0.487
4.24 2 2w1w 178 1.333
3.72 2 2w2w 465 0.972
3.19 2 2w3w 160 0.908
3.50 2 2w4w 58 1.677
4.28 2 3w1w 14 0.000
4.00 2 3w2w 51 0.998
3.45 2 3w3w 85 1.016
3.86 2 3w4w 38 0.410

3.94 3 1w1w 130 0.776
4.21 3 1w2w 70 0.581
3.20 3 1w3w 53 1.269
3.35 3 1w4w 17 1.176
4.05 3 2w1w 148 0.927
4.04 3 2w2w 476 0.789
3.42 3 2w3w 178 0.900
3.98 3 2w4w 73 1.100
4.09 3 2w5w 11 1.154
4.66 3 3w1w 30 0.563
3.77 3 3w2w 75 0.682
3.94 3 3w3w 137 0.745
4.10 3 3w4w 60 0.454
3.47 3 4w3w 19 0.768
4.46 3 4w4w 13 0.876

4.24 4 1w1w 66 0.374
4.35 4 1w2w 42 0.888
4.87 4 1w3w 16 0.000
4.51 4 2w1w 103 0.680
4.18 4 2w2w 440 0.661
3.79 4 2w3w 162 0.648
4.00 4 2w4w 46 0.739
4.64 4 2w5w 14 0.221
4.61 4 3w1w 57 0.324
4.08 4 3w2w 94 0.427
3.97 4 3w3w 217 0.718
4.19 4 3w4w 82 0.541
4.57 4 3w5w 19 0.000
5.33 4 4w1w 12 0.000
3.84 4 4w2w 26 0.580
4.12 4 4w3w 49 0.512
4.68 4 4w4w 25 1.323

4.65 5 1w1w 44 0.531
3.93 5 1w2w 31 1.380
3.73 5 1w3w 15 0.000
4.36 5 2w1w 77 0.799
4.55 5 2w2w 387 0.516
3.53 5 2w3w 101 0.873
4.06 5 2w4w 43 1.162
4.87 5 3w1w 63 0.199
4.12 5 3w2w 114 0.596
4.09 5 3w3w 239 0.729
4.37 5 3w4w 77 0.555
4.64 5 3w5w 17 0.270
4.50 5 4w1w 22 1.054
3.58 5 4w2w 41 1.136
4.19 5 4w3w 94 0.452
4.49 5 4w4w 55 0.423
3.53 5 5w3w 13 0.307

4.37 6 1w1w 24 1.270
4.44 6 1w2w 18 0.000
3.98 6 2w1w 51 1.111
4.95 6 2w2w 315 0.491
3.89 6 2w3w 106 1.047
3.74 6 2w4w 31 0.932
4.16 6 3w1w 55 1.369
4.28 6 3w2w 108 0.896
4.36 6 3w3w 215 0.777
4.49 6 3w4w 61 0.693
4.88 6 3w5w 18 0.616
5.41 6 4w1w 24 0.137
4.73 6 4w2w 67 0.420
4.75 6 4w3w 101 0.918
4.90 6 4w4w 81 0.522
5.50 6 4w5w 18 0.311
5.21 6 5w2w 19 0.436
5.22 6 5w3w 22 0.000
4.26 6 5w4w 19 0.000

4.28 7 1w1w 14 0.285
5.29 7 2w1w 24 0.204
5.26 7 2w2w 273 0.452
4.07 7 2w3w 67 0.558
4.45 7 2w4w 22 1.149
5.50 7 3w1w 32 1.462
4.33 7 3w2w 51 0.762
4.50 7 3w3w 158 0.742
4.13 7 3w4w 61 0.737
4.78 7 4w1w 28 0.374
3.90 7 4w2w 43 1.202
4.70 7 4w3w 86 0.562
5.54 7 4w4w 75 0.020
5.50 7 4w5w 22 0.195
5.25 7 5w1w 12 0.000
5.85 7 5w2w 21 0.085
4.75 7 5w3w 41 1.770
5.58 7 5w4w 34 0.185

5.17 8 2w1w 23 2.691
5.53 8 2w2w 129 0.682
4.04 8 2w3w 48 0.704
4.81 8 2w4w 11 0.509
6.38 8 3w1w 21 0.000
5.94 8 3w2w 19 0.121
5.34 8 3w3w 76 0.682
3.70 8 3w4w 27 1.074
4.57 8 4w1w 19 0.352
5.75 8 4w2w 24 0.358
5.70 8 4w3w 61 1.308
5.04 8 4w4w 49 0.238
4.73 8 4w5w 15 2.733
5.18 8 5w2w 16 1.000
3.90 8 5w3w 21 1.699
5.54 8 5w4w 11 0.000

5.84 9 2w2w 57 1.028
4.66 9 2w3w 21 0.885
4.76 9 3w2w 13 0.261
5.81 9 3w3w 33 0.163
6.86 9 4w2w 15 0.119
5.52 9 4w3w 25 0.303
5.95 9 4w4w 23 0.900
5.85 9 5w3w 14 0.142
6.00 9 5w4w 11 0.000

6.68 10 2w2w 25 0.327

Note. Count must have been greater than ten.

O.K. No more hints, just in case I do decide to play Laurel (lol).

Que.

[hdcper]

Que,

Like always your data analysis is fantastic and thanks for sharing.

Would you mind also reflecting both post 1 and 2, but show post time odds displayed and lower?

Thanks,

hdcper

[Que]

What the heck, one more table, this time grouped by the horse's prerace running style--same track and distance.


Count Post Style AvgFinish ROI

116 1 E 3.59 0.646
91 1 EP 3.89 0.925
56 1 P 3.25 1.210
313 1 PS 3.47 0.944
201 1 S 3.40 0.980
181 1 SS 3.55 1.161
428 1 U 3.70 1.081

102 2 E 3.65 1.234
90 2 EP 3.61 0.698
75 2 P 3.24 1.253
282 2 PS 3.79 1.012
178 2 S 3.89 0.839
217 2 SS 3.82 1.065
445 2 U 4.04 0.945

136 3 E 3.71 0.959
78 3 EP 3.91 1.406
49 3 P 3.77 0.808
295 3 PS 3.95 0.795
191 3 S 3.68 0.902
199 3 SS 3.88 0.694
442 3 U 4.21 0.761

98 4 E 3.96 0.528
88 4 EP 4.61 0.543
69 4 P 3.86 0.889
316 4 PS 3.93 0.746
183 4 S 4.03 0.691
186 4 SS 4.11 0.574
442 4 U 4.57 0.504

92 5 E 4.42 0.666
89 5 EP 3.85 1.013
61 5 P 3.95 0.527
307 5 PS 3.89 0.817
178 5 S 3.90 0.545
204 5 SS 4.67 0.488
435 5 U 4.56 0.565

86 6 E 4.77 1.334
55 6 EP 4.32 1.492
57 6 P 4.57 0.366
291 6 PS 4.59 0.455
183 6 S 4.51 0.575
189 6 SS 4.23 1.076
415 6 U 4.71 0.692

73 7 E 5.00 0.763
68 7 EP 5.17 0.400
42 7 P 4.73 0.814
203 7 PS 4.88 0.550
131 7 S 4.87 0.445
134 7 SS 4.84 0.416
373 7 U 4.64 0.817

31 8 E 5.25 0.419
30 8 EP 5.90 0.230
12 8 P 4.58 0.933
119 8 PS 5.11 0.994
83 8 S 5.14 1.019
74 8 SS 4.91 0.583
237 8 U 5.40 0.614

13 9 E 5.84 0.176
14 9 EP 5.57 0.242
11 9 P 6.54 0.000
52 9 PS 5.25 0.901
20 9 S 5.25 0.245
38 9 SS 5.55 1.410
119 9 U 5.73 0.404

3 10 E 6.00 0.000
3 10 EP 5.66 0.000
1 10 P 9.00 0.000
20 10 PS 5.50 0.254
8 10 S 5.12 1.112
15 10 SS 6.33 0.000
47 10 U 6.44 0.174

Now with all the above data, I'm sure I have everyone thoroughly confused--including myself.

Que.

[Que]

hdcper,

Post Position 1 -- Post Time Odds & below (http://www.100megsfree.com/que/Picture8.html)

Post Position 2 -- Post Time Odds & below (http://www.100megsfree.com/que/Picture9.html)

Hmmm... maybe those 3-5 shots in post one don't look like such a bad bet afterall.

Que.

[hdcper]

Thanks Que!

hdcper

[Dave Schwartz]

Que,

From looking at the data, would you say that it is low, mid, or high prices that are fueling the profit?

Dave

PS: Thanks for the in depth work.

[hurrikane]

well Que, great data as usual. I'm guessing you may have dropped the avg odds down to 1-5 passing this stuff around. Seems that still shows a profit though.

Curious. How does this stand up in relation to all tracks pp wins at this distance.

[andicap]

Originally posted by Rick Ransom
Dave,

Is there a logical reason for the bias at this distance at Laurel? I'm skeptical when I can't explain things like that and tend to think that things will swing the other way in the future.

I'll take a swing at it, Rick. I'm guessing this is an inherent bias at Laurel, possibly caused by the tight turns and the route races starting close to the turns.
You go 3-wide around that first turn and you're going a much longer distance and expending a ton of energy. Going wide around the first turn when horses are running faster will hurt more than going wide on the 2nd turn, I would guess. There's also a momentum factor around the 2nd turn -- horses are making a move (in relation to rest of the field, we all know it's an illusion).

[Rick Ransom]

andicap,

So the gate for 1 1/16 races at Laurel is close to the first turn? That would make sense. My point was that if they had an adequate distance before the first turn, I don't see how post position could affect things that much. Well, maybe one other way. I've heard that at some tracks where they race out of a chute, the 1 post is at a disadvantage because he's intimidated by the temporary barriers they put across the track.

Que,

Where do you get data on how many paths wide a horse is on the turns?

[Lefty]

How about a study of ELP at one mile races? The way the gate's positioned close to the turn it seems 1-2-3
posts should have the advantage.

[hurrikane]

Well Que, you think anyone sees the value of a database now?

[ridersup]

Lefty

The configuration of Ellis Parks 1 mile set up is unlike any I have ever seen. From the drawing in the DRF it looks like it starts in the middle of the clubhouse turn.

Yesterday a horse in the 11 hole was much the best in the last race at I mi but all he could manage was a 3d. A playable 7-1 shot won by many lengths from the 1 hole. The seventh race was a 1mi race and it finished 1.2.3 paying a $75 exacta.

I was set to declare a post position bonanza but the first race at 1 mi today was won by the outside 8 horse at huge odds. The other 1 mi race on the card was won by the 2 an odds on favorite.

Would rely love to see some data base work-ups on this unusual post.


Ridersup

[Rick Ransom]

In the book "Winning at the Races", Ellis Park is described as having a chute "perpendicular to the backstretch" where they start their one mile races. The diagram shows the chute on the right (clubhouse) side. Quirin's small sample shows 20.7% winning from post position 1.

[Que]

Sorry, I can't answer everyone's questions just now, since I need to start packing for a trip tomorrow morning. However, I did take a quick look at one mile routes at Ellis Park (below):


Ellis Park -- One Mile (Dirt)
Count Post AvgFin AvgOdds ROI
428 1 4.05 11.65 0.679
428 2 4.11 13.44 0.947
428 3 4.49 14.87 0.829
428 4 4.45 12.21 0.742
427 5 4.60 12.98 0.560
409 6 4.67 15.39 0.697
343 7 5.06 16.78 0.478
245 8 5.19 17.02 0.905
141 9 5.81 22.30 0.909
71 10 6.25 33.87 0.885
39 11 5.71 23.16 0.705
14 12 7.28 32.39 0.299

Just glancing at the data, it looks like post one has the best average finish, but a terrible ROI--i.e. a lot of people must like to bet the one horse. However, the number two post looks like a promising place to start.

Dave,

I don't think the post one odds were unduly influenced by any high odds horses--the distribution seemed weighted more towards the low odds horses; however, it looks like the number two post might have been influenced by some high odds horses though.

hurrikane,

I haven't looked at all the tracks, since I generally treat post position bias as track specific. However, I think it would be fairly similiar to other tracks with similiar shapes and sizes.

Anyway, got to go....

Que.

[Dave Schwartz]

Que,

And where are you going?

Is it a va-va-vacation? I had one of those once. <G>

Have a great time.


Dave

[ridersup]

QUE

Thanks a lot for the quick reply. Hope your trip is a pleasant one.


Ridersup

[Rick Ransom]

My general impression of post position bias is that whatever posts have an apparent advantage one year tend to be way overbet the next year. I think this is because there is a lot of information published for the previous year.

Some of the advantages are real and will continue the next year, but the ROI will suffer. Some of them are just random occurrences and will be bet down but the advantage will also disappear, so they will be big losers the second year. So you could either make sure the bias exists for at least two years, identify a random occurence and bet against it, or ignore the whole thing altogether, which is what I do.

Of course, if you're really quick to identify a real bias and take advantage of it, it could be worthwhile to follow it. But be careful, it won't last long. When these things are extreme, they start getting mentioned publicly by the analysts at the track.

[Lefty]

Thanks, Que. You are prob. on vacash now and I hope
you're having a great time.
Yest. at ELP had a 1mi race where I liked the 9 a lot. Oh,
my, a bad post. I didn't like the 1-2-3 at all liked the 4
2nd best. The 9 drifted up to 10-1 before I bet. Odds
overcomes all, I said to my self and bet the 9. The payoff
was over 20. I blve it was ELP 7. Sucha bad memory i'm
saddled with these days. Anyway, if you like a horse at
good odds don't get bogged down in the minutia.

[hurrikane]

Rick, in Picking Winners by Beyer and Davidowitz book they both mention the Md tracks, tight turns and bias. That was over 20 years ago. Even in the Preakness the idiot anouncers that don't know which end of the horse to stick a carrot in mentioned the tight turns and outside post disadvantage. Still...as you can see by the data...people still bet against the bias and it is still a positive play.

I can't explain it. Maybe after all this time...people get beat once or twice and they don't believe it is there anymore.

[JimL]

Consider this; eight horses are going to the gate at Pimlico with the lone speed on the outside and the Beyer figs that says he can win this race. Are you going to consider the plodder in post 1 just because of the bias? All I am saying is that the only bias that matters to me is the bias that that is inherent in the race I am working on now.

[andicap]

Here's another wrinkle which I saw at Pimlico a couple of years ago.
Speed was winning everything that day, sprint and route (the biases are usually different). We were using Multicaps and the early horses were winning even when they figured to die.
Well, by the 6th race, all the jocks had figured that out and in a route (starts near the turn), they all gunned their horses for the lead and rail. Of course they used up a lot of energy going for the lead, some got hung out wide on the first turn, big duel developed and all the speed died. A closer with a good figure two races back won and paid $20.
Now the pace bias set up for the closer (lots of speed), but earlier races which had set up for closer had also gone to front-runners which really led us all to believe there was a huge bias. Now a closer wins!
So now everyone says, "what bias: a closer just won!"
But the bias was still there and in the next race, wire-to-wire won again.

[Rick Ransom]

andicap,

I need to put that on my list of things to do ... talk to all the jockeys before the race and then I'll know how the pace will set up. Seriously though, they may make some mistakes but they're very quick to learn from them. I think they're more aware of a bias than you or I. When I see a jockey stay off the rail intentionally, I'll definitely take note of it. Similarly, if all of the jockeys are trying to get their horses up close early, I think you can be sure that there is an early speed bias. They'll realize that a closer can't win under those circumstances unless he runs faster early.

[andicap]

Rick,
I'm not sure you grasped my point --or if I grasped yours :)

I was only saying that taking advantage of a bias isn't all that easy because if the jocks all believe there's a bias and go hell bent for the lead it will lead to an early speed duel that will kill them off and set it up for a closer. Now that can also depend on the intensity of the bias. If it's a "10" the speed might still hold up. If it's an "8", maybe it won't.

Conversely if its perceived as speed dying, the jocks might try and rate their horses and a longshot early horse that gets allowed to go 4 f in 48 could hold on and win (or as a longshot, get second and key a nice exotic)

[Tom]

Originally posted by JimL
Consider this; eight horses are going to the gate at Pimlico with the lone speed on the outside and the Beyer figs that says he can win this race. Are you going to consider the plodder in post 1 just because of the bias? All I am saying is that the only bias that matters to me is the bias that that is inherent in the race I am working on now.

I would consider the rail horse if if had a logival chance, even if it were my third or fourth choice. Being outside compromises the Beyer horse somewhat, and maybe the rail bias move the "plodder ahead a length-maybe enough on both sides to warrant the rail horse. Yu have to use bias information intelligently as a handicapping factor, and determine how much to weigh it. But then again, just becasue a long term bias exists, you aren't betting the average, you are betting a specific race and you have to acknowledge the race-bias as well as the track bias. My experrience is that the value of a track bias is usually over-played and more percieved than real. I think only a few actual races can tell you if there is a real bias. When a horse figures to quit and doesn't, when the 1-2-3 horses at the FCP are the 1-2-3 finishers, when horses out-running their odds all share a running trait or post, then you might have one.
The best bias is the one that doesn't exist but everyone is thinking it does. When a 3-5 shot wires the fisrt race, and the 4-5 shot wires the second, 80% of the people at the track that are still sober or conscious will say"Speed is golden! Yea!"
Tom

[Rick Ransom]

andicap,

No, I got your point and I agree. It's just that I can't resist a chance to be sarcastic. Just part of my personality.

[andicap]

Oh, so Rick was being sarcastic. Is he overwhelming us with his wit? I wish I could be as funny as Rick.

[Rick Ransom]

andicap,

Ok, I get it. Turnabout is fair play! Congratulations, my friend.