Speed Maps / Profiling – Nature or Nurture?

There is often debate over how speed maps are best constructed and what information should be used as the basic building blocks. In fact, the use of terms such as speed maps and speed profiles has connotations that if you use some form of raw early speed information, you will end up with the most reliable representation of predicted position in running (or settle position). Unfortunately, such debate often comprises confident statements not supported by any form of rigorous analysis. Punting Form (in association with ABettorEdge) have two types of measures to look at (both developed in-house) in attempting to determine whether horses take up running positions according to their raw early speed ability or more in keeping with their historical running style. We will only consider sprint races up to and including 1200m in distance as part of our analysis.

The first measure shown in the table below, displays a predicted settle position using only historical settle position data as the basis for the prediction, versus actual settle position. Those predicted to lead in the race did so on 29.3% of occasions and were in the first 3 in running 62.6% of the time. This was from a sample of approx. 1,800 races.

 

PrSe

Predicted Settle Position (PrSe) Vs Actual Settle Position (SePos)

Count of SePos

% of Total

Cumulative % of Total

1

SePos

1878

100.0%

 

 

1

551

29.3%

29.3%

 

2

351

18.7%

48.0%

 

3

274

14.6%

62.6%

 

4

171

9.1%

71.7%

 

5

140

7.5%

79.2%

 

6

101

5.4%

84.6%

 

7

79

4.2%

88.8%

 

8

61

3.2%

92.0%

 

9

41

2.2%

94.2%

 

10

35

1.9%

96.1%

 

11

27

1.4%

97.5%

 

12

17

0.9%

98.4%

 

13

18

1.0%

99.4%

 

14

6

0.3%

99.7%

 

15

5

0.3%

99.9%

 

16

1

0.1%

100.0%

 

Our second measure, shown in the table below shows a predicted settle position using only raw early sectional time speed data (ETRPrR) as the basis for the prediction, versus actual settle position. Those predicted to lead in the race now did so on 25.1% of occasions and were in the first 3 in running 55.1% of the time. This was from the same sample of approx. 1,800 races.

 

ETRPrRk

Early Speed Rating Rank (ETRPrRk) Vs Actual Settle Position (SePos)

Count of SePos

% of Total

Cumulative % of Total

1

SePos

1876

100.0%

 

 

1

470

25.1%

25.1%

 

2

325

17.3%

42.4%

 

3

238

12.7%

55.1%

 

4

164

8.7%

63.8%

 

5

148

7.9%

71.7%

 

6

121

6.4%

78.1%

 

7

112

6.0%

84.1%

 

8

80

4.3%

88.4%

 

9

76

4.1%

92.4%

 

10

43

2.3%

94.7%

 

11

42

2.2%

97.0%

 

12

25

1.3%

98.3%

 

13

15

0.8%

99.1%

 

14

11

0.6%

99.7%

 

15

3

0.2%

99.8%

 

16

3

0.2%

100.0%

 

The conclusion we can draw from this is that 'historical position in running' is more dominant as a predictor of settle position in running for an upcoming race, than 'raw early speed ability' demonstrated in previous races (29.3% versus 25.1%). If raw early speed was the only (or primary) contributor, then the two measures investigated should have delivered similar results.

So, there is something else at play and I would suggest that it is either a 'herd' instinct or 'learned running style' as a result of training patterns. In keeping with not wanting to assert assumptions as fact (without accompanying analysis), we will leave it at that for the time being.

What we can do is look at how the two measures interact with one another. It makes sense that a runner with top ranking on historical run-style and the top ranking on early speed ability should lead races at a higher rate than either factor considered independently. This in fact is the case, with this type of runner leading on 43% of occasions (up from 29% based on one measure only). Likewise, a runner with top ranking on historical run-style and a ranking on early speed ability outside of the top 6, ends up leading races at a rate of 13%, much lower than the single measure, 29% benchmark.

Knowing both the ranking of a horse based on historical run-style (available in the Punting Form form-guides) and raw early speed ability (available in the Desktop Ratings software) in today's race, is potentially very useful in determining how much energy a horse is likely to expend in trying to attain its preferred running style. Those with top billing on both measures will be required to expend very little early energy, while those who like to lead but don't have the necessary  raw early speed ability, will expend a lot of energy struggling to take up the front running position.

Punting Form has Speed Maps based on historical run style freely available for the majority of thoroughbred races run in Australia. By becoming a registered user (a free process) you can also drag and drop runners from their default positions, thereby creating and saving your own personalized race speed maps. An example speed map is shown below:

 

Speedmapshot

Comments

croc posted this comment 01/23/2014 10:43:12 AM

Great article. I'm not exactly sure what your formula is, however if it is only using the previous settle positions as you have said, then that is probably not ideal. Poor past performances don't offer as much insight into future settle positions. I regard a poor past performance as one where the horse performed below expectations. Incorporating previous SP, finishing margin and the settle position may offer increased predictive ability. I also believe the Jockey's ability and the barrier is paramount to where a horse settles and ultimately finishes, so they are also hard to ignore. So now I'm at 5 factors from various past performances to predict a single outcome for today's race, that's very difficult to do with accuracy. And then what do I do with this new prediction? Can I use it to calculate winning chances more accurately? Sorry for sounding like a debbie downer, but I sometimes I think that the more you know the harder it becomes to make a profit. Nowadays all I ever hear on the radio each raceday is where a horse may or may not settle and the pace etc etc. It's like an obsession. And they seem to ignore or under estimate the importance of the most highly predictive factors such as Recent form, Fitness, Ability, Jockey.

ABettorEdge posted this comment 01/31/2014 16:06:24 PM

Thanks for taking the time to leave a comment. I think most of your points are valid from a form-study point of view. The thing to remember here is that for both cases under examination (Predicted Settle based on Historical Run Style Vs Early Speed Ability) we have the same pool of horses contesting the same races. This means any 'other' circumstances such as those you raise, do not have any impact on the comparison made and inferences drawn.

bsmith9902 posted this comment 01/31/2014 18:13:26 PM

Interesting read. Keep them coming!

croc posted this comment 02/02/2014 07:50:46 AM

Of course you 100% correct, your study was to compare the predictive ability of 2 factors, raw early speed and historical running positions. And the conclusion that historical running positions have greater predictive power is an important one. Also combining the two clearly shows that both have some independent predicative power over each other.