Finish Position/Final Odds Comparison

[traynor]

In 15888 Pace Races Average Odds for First were 3.63
In 15888 Pace Races Combined Odds for First were 10.18
In 15888 Pace Races 66.39 % Lower 10.74 % Equal 22.87 % Higher 55.38 % Over 2-1 This 21.85 % Under 1-1 This
In 15888 Pace Races Purse 28.86 % Lower 54.56 % Equal 16.58 % Higher

In 18385 Pace Races Average Odds for Second were 5.70
In 18385 Pace Races Combined Odds for Second were 13.07
In 18385 Pace Races 52.73 % Lower 17.76 % Equal 29.50 % Higher 78.21 % Over 2-1 This 7.05 % Under 1-1 This
In 18385 Pace Races Purse 24.75 % Lower 57.31 % Equal 17.94 % Higher

In 19370 Pace Races Average Odds for Third were 7.29
In 19370 Pace Races Combined Odds for Third were 15.23
In 19370 Pace Races 40.81 % Lower 26.45 % Equal 32.74 % Higher 88.08 % Over 2-1 This 3.19 % Under 1-1 This
In 19370 Pace Races Purse 22.89 % Lower 57.87 % Equal 19.24 % Higher

In 20050 Pace Races Average Odds for Fourth were 8.28
In 20050 Pace Races Combined Odds for Fourth were 16.41
In 20028 Pace Races Average Odds for Fifth were 9.07
In 20028 Pace Races Combined Odds for Fifth were 17.56

In 19941 Pace Races Average Odds for Sixth were 9.66
In 19941 Pace Races Combined Odds for Sixth were 18.33

In 18350 Pace Races Average Odds for Seventh were 10.09
In 18350 Pace Races Combined Odds for Seventh were 18.96

In 13822 Pace Races Average Odds for Eighth were 10.26
In 13822 Pace Races Combined Odds for Eighth were 19.26

In 7436 Pace Races Average Odds for Ninth were 10.41
In 7436 Pace Races Combined Odds for Ninth were 19.38

Clearly, chasing chalk selections is not useful if one is intent on making a profit. "Combined Odds" are last race odds + current race odds. This also makes the "wisdom of crowds"--if the crowd wagering on harness races is considered--look more like the "foolishness of crowds."

[traynor]

What "21.85 % Under 1-1 This" means is that the expected return (best case scenario) on wagers on entries going off at odds below even money is $87.20 in each 100 races so wagered, at a cost of $200, for an ROI of 0.436.

Secondarily, it means that models projecting win probability and expected ROI can be substantially improved by tossing the chalk. The problem is that most want to include the chalk (to fluff up the win%) without realizing how seriously that affects profitability.

Unfortunately, the races that (when studied as past events) indicate a high percentage of winners are the very races that drag down the ROI. Toss the chalk, up the actual ROI on the same races selected using the same methods. Of course, it is essential to clean the data of outliers and layer appropriately so one is not chasing rainbows. Meaning that the win percentage after culling outliers and eliminating chalk races, multiplied by average mutuel, must be high enough to be profitable.

[mrroyboy]

Good research and nice to see you posting again Tray
I don't really understand the charts but I sort of understand your explanations.
Hey, I have trouble figuring out Facebook!!

[traynor]

Originally Posted by mrroyboy

Good research and nice to see you posting again Tray
I don't really understand the charts but I sort of understand your explanations.
Hey, I have trouble figuring out Facebook!!

A lot of (unstated) information is contained in the figures. Small examples:
"In 15888 Pace Races Purse 28.86 % Lower 54.56 % Equal 16.58 % Higher" means that in 83.42% of the races, the winner was NOT dropping in purse value from its last race. That pretty much blows out the idea that "dropping in class" is some kind of advantage. The overwhelming majority of such "class droppers" lose.


Similarly, "In 15888 Pace Races 66.39 % Lower 10.74 % Equal 22.87 % Higher" means that in 77.13% of the races, the winner went off at odds equal to or LOWER than in its preceding race. That pretty much blows out the idea that "beaten favorites" that had a bad trip (or some other excuse) are worth betting the next time out. That is only true IFF ("if and only if") the odds of that entry in the current race are equal to or less than the odds in the preceding race. Looking for "value wagers" in such races can be costly.

[mrroyboy]

Thank you Tray

[DRIVEWAY]

These statistics are very interesting. Curious to know what the percentages would be if all two year old races were ignored.

[traynor]

Originally Posted by DRIVEWAY

These statistics are very interesting. Curious to know what the percentages would be if all two year old races were ignored.

Its a good question, but because I usually don't consider betting on two-year-olds, and there is little value in layering by age, I don't consider age in the generic figures. The differences would be more evident in select races (a category) that discards all two-year-old races.

[traynor]

Same situation with odds-on entries in trot races. Best case scenario is a return of $80.84 for $200 bet for an ROI of a whopping 0.4042.

Similarly, class droppers did not fare well. Only 17.57% of the winners had been entered at a higher purse level in the preceding race.

In 6734 Trot Races Average Odds for First were 3.78
In 6734 Trot Races Combined Odds for First were 10.19
In 6734 Trot Races 63.10 % Lower 12.53 % Equal 24.37 % Higher 58.21 % Over 2-1 This 20.21 % Under 1-1 This
In 6734 Trot Races Purse 29.91 % Lower 52.52 % Equal 17.57 % Higher

In 7862 Trot Races Average Odds for Second were 5.89
In 7862 Trot Races Combined Odds for Second were 13.13
In 7862 Trot Races 49.17 % Lower 18.40 % Equal 32.42 % Higher 79.38 % Over 2-1 This 6.16 % Under 1-1 This
In 7862 Trot Races Purse 26.10 % Lower 54.40 % Equal 19.50 % Higher

In 8055 Trot Races Average Odds for Third were 7.51
In 8055 Trot Races Combined Odds for Third were 15.41
In 8055 Trot Races 38.46 % Lower 27.14 % Equal 34.40 % Higher 88.73 % Over 2-1 This 2.95 % Under 1-1 This
In 8055 Trot Races Purse23.53 % Lower 54.49 % Equal 21.99 % Higher

In 8299 Trot Races Average Odds for Fourth were 8.51
In 8299 Trot Races Combined Odds for Fourth were 16.89

In 8131 Trot Races Average Odds for Fifth were 9.28
In 8131 Trot Races Combined Odds for Fifth were 17.82

In 7771 Trot Races Average Odds for Sixth were 9.65
In 7771 Trot Races Combined Odds for Sixth were 18.43

In 6463 Trot Races Average Odds for Seventh were 9.89
In 6463 Trot Races Combined Odds for Seventh were 18.68

In 4510 Trot Races Average Odds for Eighth were 10.17
In 4510 Trot Races Combined Odds for Eighth were 19.00

In 2443 Trot Races Average Odds for Ninth were 10.10
In 2443 Trot Races Combined Odds for Ninth were 18.76

[mrroyboy]

Tray you are using purse value for class instead of TM class numbers right?

[imofe]

Originally Posted by traynor

Same situation with odds-on entries in trot races. Best case scenario is a return of $80.84 for $200 bet for an ROI of a whopping 0.4042.

Similarly, class droppers did not fare well. Only 17.57% of the winners had been entered at a higher purse level in the preceding race.

In 6734 Trot Races Average Odds for First were 3.78
In 6734 Trot Races Combined Odds for First were 10.19
In 6734 Trot Races 63.10 % Lower 12.53 % Equal 24.37 % Higher 58.21 % Over 2-1 This 20.21 % Under 1-1 This
In 6734 Trot Races Purse 29.91 % Lower 52.52 % Equal 17.57 % Higher

In 7862 Trot Races Average Odds for Second were 5.89
In 7862 Trot Races Combined Odds for Second were 13.13
In 7862 Trot Races 49.17 % Lower 18.40 % Equal 32.42 % Higher 79.38 % Over 2-1 This 6.16 % Under 1-1 This
In 7862 Trot Races Purse 26.10 % Lower 54.40 % Equal 19.50 % Higher

In 8055 Trot Races Average Odds for Third were 7.51
In 8055 Trot Races Combined Odds for Third were 15.41
In 8055 Trot Races 38.46 % Lower 27.14 % Equal 34.40 % Higher 88.73 % Over 2-1 This 2.95 % Under 1-1 This
In 8055 Trot Races Purse23.53 % Lower 54.49 % Equal 21.99 % Higher

In 8299 Trot Races Average Odds for Fourth were 8.51
In 8299 Trot Races Combined Odds for Fourth were 16.89

In 8131 Trot Races Average Odds for Fifth were 9.28
In 8131 Trot Races Combined Odds for Fifth were 17.82

In 7771 Trot Races Average Odds for Sixth were 9.65
In 7771 Trot Races Combined Odds for Sixth were 18.43

In 6463 Trot Races Average Odds for Seventh were 9.89
In 6463 Trot Races Combined Odds for Seventh were 18.68

In 4510 Trot Races Average Odds for Eighth were 10.17
In 4510 Trot Races Combined Odds for Eighth were 19.00

In 2443 Trot Races Average Odds for Ninth were 10.10
In 2443 Trot Races Combined Odds for Ninth were 18.76

I believe that droppers produce more overbet horses than just about anything. I am not talking about a horse that drops to a condition that he just raced at in the prior start. I am talking about a horse that has all lines at a higher condition. I can hear the guy next to me at the track spouting off the list of winners this horse has been in against. If the horse is not competitive at all in this higher class, he is not the CLASS of the race and his time and/or speed rating is higher than it should be because of just following along faster horses.

[mrroyboy]

Originally Posted by imofe

I believe that droppers produce more overbet horses than just about anything. I am not talking about a horse that drops to a condition that he just raced at in the prior start. I am talking about a horse that has all lines at a higher condition. I can hear the guy next to me at the track spouting off the list of winners this horse has been in against. If the horse is not competitive at all in this higher class, he is not the CLASS of the race and his time and/or speed rating is higher than it should be because of just following along faster horses.

Agree those are suck a long races. That's the problem with TM average class ratings.

[Ray2000]

Are you sure the return is on $200? Those ROIs look like they should be doubled.

[imofe]

Originally Posted by mrroyboy

Agree those are suck a long races. That's the problem with TM average class ratings.

It is a good thing. That is what creates the illusion of the horse being a contender. Speed rating is high and class rating is high. Most people are afraid not to use a horse like this.

[traynor]

Originally Posted by mrroyboy

Tray you are using purse value for class instead of TM class numbers right?

I don't use TM class numbers at all.

[traynor]

Originally Posted by Ray2000

Are you sure the return is on $200? Those ROIs look like they should be doubled.

Odds < 1/1, wins 20.21 races of 100 races, best case scenario each win returns $4. $4 (20.21) = $80.84/200 (cost of the 100 $2 bets). ROI 0.4042.