False Favorites

[MikeDee]

Did a little reaserch on false favorites that looks interesting. I took the factors of Pow, Per, K, Htr, pscan and mike# and added the rankings. The best you could be is 6 (6*1) and the worst you could be is 54 (9*6).

Then I filtered on MLR (morn line rk) = 1 and off odds rank =1 and the sum of the above 6 factors >=18 (this would be an average of a rank of 3 for each of the 6 factors. Looked at all races on fast dirt excluding races with any first time starters and here is what I got.
Trk nWin nBet Win% aMut
AP 17 80 21% $5.59
BEL 23 78 29% $5.26
CRC 21 103 20% $5.57
CT 20 76 26% $6.45
DEL 22 99 22% $5.68
FE 27 132 20% $5.67
KEE 8 43 19% $4.88
LAD 10 67 15% $6.40
MNR 34 118 29% $6.15
PEN 17 96 18% $5.59
PHA 45 162 28% $5.64
PRM 13 84 15% $5.77
RET 2 22 9% $5.00
SA 17 79 22% $5.35
TDN 47 176 27% $5.51
TP 31 138 22% $5.35
WO 48 198 24% $5.29

I used both MLR and Off odds rank based on the concept that the public confirms the MLR by keeping it the fav while the HTR factors don't.

It looks like the ML handicapper and the public are pretty good at BEL, MNR, PHA, and TDN, but at the other tracks the fav doesn't do so good and maybe these are races where looking for overlays could payoff.

I would expect that field size could drop this win% still more if you filtered on fields of >=8 but didn't test it, would also reduce the number of plays.

Didn't check the reverse which would be fav's = 6 mlor =1 off odds=1 but checked it at on AP and it was 41%

[MtKen]

MD:

Good stuff. PL 5?

What kind of odds do you give the Tribe to win the AL? Think I'd have to have 12/1 to make a bet. Yanks & Mariners around 2/1 to 3/1 & A's about 6/1.

[MikeDee]

yep PL5

never was any good at setting a odds line. I want more then 12-1 more like 20-1.

Haven/t done very good at all with Sea or Oak

[Sam Thorp]

Mike: how do you make Access add the six rankings and test that number?

[MikeDee]

take any query you have. If it is a new one save it first. Then go to the design view take any empty column put your cursor in the top row, where the field name would go and do a right click on the pop up menu select build. this will open screen where you can calculate a field.

In the second window you will see all of the fields in your query just scan down the names and click on the first one you want (like rpow) and double click it now put a "+" sign after it.
now double click on the next one and double click it. continue putting any and all fields you want. It will name your expression expr1: and you can change that to suit

[Sam Thorp]

We learn something new everyday. Thanks for the help, Mike.

My other question is this: I remember that impact has the rank of morning line odds. But I can't find where rank of Off odds is. Can somebody help with that?

[Glen]

Excellent Mike. How often do the false favorite plays come available?

[MikeDee]

Sam
There isn't a off odds rank in HTR so for the study I jsut use a vba module to rank them. If you are using this at post time as you know usually the psot time fav does'nt change to often from the no 1 ranked fav. to often. But this is a good computer play when youcan wait to the last min to wager.

Glen
good question. I will do a run on the last weeks data and give you a day by day look. There are alot of potential races that qualify but the public figures out most of them as well and they don't go off as the fav.

[MikeDee]

Ran the false fav analysis for the past week for the tracks shown on attached. This is same filters as stated in the first except all races on "dirt or wet" and fields >6. Overal results are:

60 opportunities 14 winners for 23.3% win rate for qualifing favorites. The attached worksheet shows the number of opportunities by track by day. My week runs from Monday thru Sunday

[MtKen]

Mike:

Great stuff again! Anyway to EASILY figure out the win prices that came out of these races?

[MikeDee]

Yea it would have been easy but I don't have all the data anymore. Was just checking it as a by-product of my weekly updates. Will check the tracks I follow and post them tomorrow.

Today at del, crc, fe and lad there were 3 qualifing races, each of the fav's lost and the winners paid 9.80, 9.00 and 9.40. I had a $45 exa in one of them (just practicing for Wednesday):)

[Gramps]

Then one has to learn VBA programming to make these modules?

[Sam Thorp]

Rick: is it possible to get a copy of that module?

[MikeDee]

Gramps

- Are you back or lurking?

There are some things that I couldn't figure any other way to do other then write a program for. If you have some time and the inclination for it, learning how to write simple basic programs in Access is not a big deal. Things like creating compound values from existing values and ranking these new values from 1 to n. Besides I've already done this kinda stuff before and I can send you the code and help you with getting it running.

Sam
I assume you are addressing your question to me. You want the module that ranks off odds correct? Send me a email (mrdezo@ameritech.net) and I'll send it to you.