Zayonce, can you explain this mathmatically? I have to program this all in and I don't really follow what you're asking.

What's Winning at GPX 2021 Dirt Sprints

If you have any questions on anything, please let me know

Okay, back from the ether.

I found a few items on time decay variables but the most informative was Jeff Pratt's (jCapper Jeff) description of some new jCapper time delay variables. He uses a half life of 120 days so I tried that. Taking the maidens out of the spreadsheet gives us a win percentage at GPX 2021 of 21.9% with an roi of 1.77.

Next question "is 120 days a good half life or is there a better one, 90 days, 75 days?????". I'll get around to answering my own question when I get to sit down in front of the computer for another couple of hours.

Dave, did you know Bill?

I've heard you speak of Time Decayed APV and I've wondered what it was. Now I will wander into the ether and seach for Benter's Time Decay and see if I understand it.

Thanks!

Thank you Dave, that's very kind of you.

This is (as well as I am) still a work in progress and I hope to add to the What Wins program/spreadsheet as I go along. I'm all for massive amounts of data, I'm just not sure what would be worth putting in there. I have added Quirin Early Speed points and I'm looking for other factors to add. I don't think I'll ever get to 500 but I am already set up for about 100 (I just need factors). Anybody got something they want to look at (low level type factor, best speed at distance or track for example would be okay. Nothing that involves too many calculations)?

I thought this was most interesting so I went back and reviewed WHAT WINS - Dave Schwartz and worked up the attached for the spring meet at Keeneland. I am not to the point where I could actually use this data but I thought that some one else might have an insight or three and would share.

Most headings I hope are self explanatory but if not.... (the _# is just to help keep me straight in the code, it's not relavent to much of anything else)

TRK_1 track name
DATE_2 date of race
RACE_3 race number
ML_10 morning line odds
FINSH_15 finish
PT_ODDS_16 post time odds
WIN_17 win payoff
PLACE_18 place payoff
SHOW_19 show playoff
ML_RANK_20 morning line odds rank
PT_RANK_21 post time odds rank
PRIMEP_RANK_22 prime power rank
EP_RANK_23 early pace (my version) rank
LP_RANK_24 late pace (my version) rank
SAR_RANK_25 some combination of Sartin fps numbers (I would have to look closely to see the computations)
B_365_RANK_26 best speed in 365 days rank
BL2_RANK_27 best of last two speeds rank
TR/JO_RANK_28 combined trainer jockey win percentage rank
LRS_RANK_29 last race speed rank

Just wanted to bump the thread.

Guilty as charged! At least to the point of not betting anything lower than 3-1.