|
Is there really a bias?
By David E. Schwartz
This question turned out to be amazingly
simple to answer. In our database, we show 1,524 races run on
fast tracks. If we
simply break them down by post position, here are the results.
 |
|
Laurel, 1 1/16 miles - Post
Position |
|
|
Starts |
Wins |
Win% |
$Net |
IV |
PIV |
Avg
Pay |
Avg
Odds |
Avg
PubCh |
|
1 |
1,524 |
352 |
23.1 |
$2.32 |
1.71 |
1.38 |
$10.03 |
4.05 |
3.7 |
|
2 |
1,524 |
264 |
17.3 |
$1.84 |
1.28 |
1.20 |
$10.62 |
4.87 |
4.1 |
|
3 |
1,524 |
221 |
14.5 |
$1.67 |
1.07 |
1.03 |
$11.50 |
5.02 |
4.1 |
|
4 |
1,522 |
182 |
12.0 |
$1.29 |
0.88 |
0.88 |
$10.81 |
5.26 |
4.3 |
|
5 |
1,511 |
168 |
11.1 |
$1.39 |
0.83 |
0.81 |
$12.51 |
5.17 |
4.2 |
|
6 |
1,460 |
144 |
9.9 |
$1.22 |
0.75 |
0.77 |
$12.37 |
5.64 |
4.5 |
|
7 |
1,214 |
94 |
7.7 |
$1.00 |
0.62 |
0.68 |
$12.89 |
6.45 |
4.7 |
|
8 |
763 |
66 |
8.7 |
$1.57 |
0.76 |
0.81 |
$18.12 |
6.91 |
4.9 |
|
up
|
623 |
39 |
6.3 |
$0.86 |
0.60 |
0.67 |
$13.67 |
8.08 |
5.5 |
|
total |
11,665 |
1,530 |
13.1 |
$1.51 |
1.00 |
0.98 |
$11.55 |
5.30 |
4.3 |
This data shows without a doubt that the 1 1/16 mile distance at
Laurel truly is beatable simply with post position.
Perhaps the most interesting thing about
this study is that the public has knowledge of this bias as evidenced
by the average odds and public choice columns. In spite of their
knowledge they simply under bet these horses. Let's
break this study down by Quirin Early Speed Points and see if
there is any pattern.
(Editor's note: The totals in the following sections will not
match the totals above because first-time starters and foreign
starters have been removed.)
|
Closers - 0 Early Speed Points |
| PP |
sts |
win |
w% |
IV |
| 1 |
317 |
48 |
15% |
1.13 |
| 2 |
317 |
42 |
13% |
1.01 |
| 3 |
312 |
34 |
11% |
0.83 |
| 4 |
326 |
34 |
10% |
0.79 |
| 5 |
312 |
25 |
8% |
0.60 |
| 6 |
273 |
26 |
10% |
0.73 |
| 7 |
254 |
10 |
4% |
0.32 |
| 8 |
177 |
16 |
9% |
0.79 |
| 9 |
114 |
8 |
7% |
0.66 |
| 10 |
54 |
2 |
4% |
0.37 |
| 11+ |
0 |
0 |
0% |
- |
| Total |
2,456 |
245 |
10% |
0.78 |
|
Mid-Pack Horses - 1 to 4 Early
Speed Points |
| PP |
sts |
win |
w% |
IV |
| 1 |
587 |
132 |
22% |
1.68 |
| 2 |
592 |
99 |
17% |
1.22 |
| 3 |
601 |
81 |
13% |
1.00 |
| 4 |
559 |
56 |
10% |
0.75 |
| 5 |
575 |
67 |
12% |
0.87 |
| 6 |
623 |
56 |
9% |
0.68 |
| 7 |
480 |
41 |
9% |
0.69 |
| 8 |
311 |
28 |
9% |
0.79 |
| 9 |
170 |
11 |
6% |
0.61 |
| 10 |
81 |
5 |
6% |
0.62 |
| 11+ |
0 |
0 |
0% |
- |
| Total |
4,579 |
576 |
13% |
0.97 |
|
Front Runners - 5 to 8 Early
Speed Points |
| PP |
sts |
win |
w% |
IV |
| 1 |
613 |
174 |
28% |
2.06 |
| 2 |
611 |
126 |
21% |
1.51 |
| 3 |
606 |
102 |
17% |
1.21 |
| 4 |
631 |
90 |
14% |
1.04 |
| 5 |
620 |
72 |
12% |
0.86 |
| 6 |
559 |
63 |
11% |
0.85 |
| 7 |
476 |
41 |
9% |
0.68 |
| 8 |
272 |
23 |
8% |
0.74 |
| 9 |
142 |
10 |
7% |
0.66 |
| 10 |
60 |
4 |
7% |
0.67 |
| 11+ |
0 |
0 |
0% |
0 |
| Total |
4,590 |
705 |
15% |
1.16 |
And what of the obvious question of track
condition?
In this sample, we consider only those races
run on "off" tracks. That is, races that were
not "fast" or "good."
| ALL |
Sts |
Pays |
PCT |
$NET |
| 1 |
169 |
47 |
27.8 |
$3.08 |
| 2 |
169 |
30 |
17.8 |
$2.29 |
| 3 |
169 |
25 |
14.8 |
$1.98 |
| 4 |
169 |
23 |
13.6 |
$1.18 |
| 5 |
162 |
16 |
9.9 |
$1.06 |
| 6 |
144 |
13 |
9.0 |
$1.22 |
| 7 |
114 |
6 |
5.3 |
$1.70 |
| 8 |
63 |
8 |
12.7 |
$2.56 |
| up |
43 |
1 |
2.3 |
$0.42 |
| Total |
1,202 |
169 |
14.1 |
$1.80 |
Here we see that the bias is about the same
from a standpoint of win%, but that the public manages to forget
to consider post position all together!
By the way, the sample of "good"
tracks was small (70 races) and showed a diminished bias.
Summary There
really is a post position bias at Laurel! It is so pronounced
that if one had done nothing but bet horses in post position one
over the last few years, a profit of at least 16% per wagered dollar
would have been realized. Furthermore,
looking at the impact of running style, using the Quirin Early
Speed Point approach, shows us that the greater the likelihood
of the horse being a front runner, the stronger the bias
becomes. On off tracks, the mid-pack horses
showed improvement over the front runners, but closers did
dismally. In other words, the bias is still intact. |
|
|
