Couple of queries and a Process for Mt. Ken.

[Carl]

It's funny how queries take on a life of their own, for me anyway. My process seems to be I ploud around till I find something unusual, then that is what I explore further.

All of which is by way of preface to what follows.

I thought I would do a break down for Ken of routes and sprints, maybe Md/non-Md. But didn't turn out that way.

So where I started was to do a breakout of the more promising of the queries I ran, the one with 3945 races that showed a plus ROI. To refresh that was:

ALL_MSA Misc Query ROI Report
All tracks July 2000 to March 2001
nESPD >=9 and rTRN >4

Total Bets 3945
Total Amount Bet 7890.00
Wins 661
Pct. 16.76%
Amount Won 8107.70
Profit/Loss 217.70
Pct Profit Loss 2.76%
Avg. Payout 12.27


So I broke this set up into sprints (<8F) and routes (>=8F).


ALL_MSA Misc Query ROI Report
nESPD >=9, dist <8, rTRN >4

Total Bets 2376
Total Amount Bet 4752.00
Wins 437
Pct. 18.39%
Amount Won 4992.40
Profit/Loss 240.40
Pct Profit Loss 5.06%
Avg. Payout 11.42



ALL_MSA Misc Query ROI Report
nESPD >=9, dist >=8, rTRN >4.

Total Bets 1569
Total Amount Bet 3138.00
Wins 224
Pct. 14.28%
Amount Won 3115.30
Profit/Loss -22.70
Pct Profit Loss -0.72%
Avg. Payout 13.91

And I got sidetracked.

This wasn't what I expected. I was "sure" routes would test better for ROI with this play than sprints.

And it didn't.

So........ I start to think "well maybe the 'bad' (>4 rTRN) trainers cause this, they can't train routers".

So my query changes direction (sorta by itself).

Now I want to know how the first set of data (with rTRN 1) I ran for George will do broken down by sprints and routes. So I go back to that set. To refresh that was:

ALL_MSA Misc Query ROI Report
All tracks July 2000 to March 2001
nESPD >=9 and rTRN 1.

Total Bets 1770
Total Amount Bet 3540.00
Wins 577
Pct. 32.60%
Amount Won 3306.40
Profit/Loss -233.60
Pct Profit Loss -6.60%
Avg. Payout 5.73

Then this set broken down by sprints and routes:

ALL_MSA Misc Query ROI Report

nESPD >=9, dist <8, rTRN 1.

Total Bets 1087
Total Amount Bet 2174.00
Wins 381
Pct. 35.05%
Amount Won 2125.20
Profit/Loss -48.80
Pct Profit Loss -2.24%
Avg. Payout 5.58


ALL_MSA Misc Query ROI Report

nESPD >=9, dist >=8, rTRN 1.

Total Bets 683
Total Amount Bet 1366.00
Wins 196
Pct. 28.70%
Amount Won 1181.20
Profit/Loss -184.80
Pct Profit Loss -13.53%
Avg. Payout 6.03

Amazing for me. This set tests better for sprints too! Who woulda thunk it? (Well, in hindsight I can, as there are only two qualifiers for this play, etc. etc. etc.)

In any case, here I am an hour into this thing, and not a heck of a lot smarter. Well, maybe. I like the play better for sprints than I did.

I am posting all of this to let everyone know that
1) queries are endless, and
2) they do direct you as well as you direct them.

Sorry I didn't get further along Ken, but the soaps are over and duty beckons....hehe.

later,
Carl

[MikeDee]

Since espeed dif of >=9 seems to be the break point of preference I decided to do a comparison of >=10 vs >=9. I have added some additional tracks to my db. I only bring in those tracks that are running and I can bet on line at UBET. Have a couple more to go.

So, due the ongoing struggle between Magna and the gang of three (CD inc, NYRA and NTRA) for the online wagering business. I don't have any NYRA tracks or CD inc tracks in my db.

Kinda of interesting dif 10 has a better win% and ROI at those tracks where the ROI is + and the tracks with a - ROI have a less negative ROI with the dif of 9.

On the dif of 10 it seems the the win% of 22% is the break point tracks that have a win% of >=22% have good ROIs while those below this point have the -ROIs.

I'm going to add BM GLD and CT to the mix and then I'm going to test some the the factors like sprint vs route, field size, trainer rank and a few others

[George]

Mike, very interesting. The answers we get are always a little different depending on db size and mix of data. Used the Actual/Expected (AE) ratio to analyze nESD plus numbers from +1 to +15 one at a time. On my 8500 race db with mostly cheap tracks the best AE was +9 followed closely by +10.

Have found the AE to be an excellent tool to quickly point out factors that have a good mix of decent win% with lesser "betting pressure". Used AE to come up with the 2 factor query we have been playing around with.

Will be interested in what you come up with for nESD as you slice/dice it.

[MtKen]

Carl:

Thanks again...certainly makes sense to follow the profit & see where it comes from. Before Impact I had pretty much limited my play to EV1's in short sprints, <=6f, & maintained a healthy Roi but had some killer losing streaks. At that point I could never figure out how people were playing routes for EV1...Now I have an Impact play for routes that is doing OK but doesn't get as much action as my sprint stuff. I also limit my route plays at 20/1 max actual odds. I spend a lot of time tote-watching & try to get my bets in as late as possible on Winticket.

Are you still betting off-shore? I seem to recall you mentioning that you had to get your larger bets in pretty early, is that right??? I've been mainly in $5-$10 range lately, down from $20 when I hot in Jan. Think they would take small bets like that with 1 or 2 mtp?

Ken

[Carl]

Ken,

I am still using ESB for AQU (can't bet it with YOUBET), and I have a Aces Gold Account which I need to fill to re-activate.

Neither of them really care what you bet, but both are time-sensitive. Aces Gold cuts off a little before scheduled post, and ESB refunded my bet once because I bet it too close to post (it was a loser, so that was nice). ESB recommends you bet 10 minutes before post to be safe, so I do.

They seem to treat you the same no matter how much you bet, which is I guess good customer service.

later,
Carl

[Carl]

Still on the trial of those missing route winners.

Thought of a new place to look. Maybe they are hanging out with the one group of trainers I haven't checked yet, those ranked 2,3,4. (I did two queries re George's post, one for nESPD >=9 with rTRN=1, one with >=9 and rTRN >4).

And sure enough, there they are. This the kind of "relationship" that drives me crazy, and why I wouldn't use neural ants if I knew how. Somethings are even too strange for me. Anyway, here are the three queries for nESPD >=9, rTRN 2,3,4.

ALL_MSA Misc Query ROI Report
nESPD >=9, rTRN 2 or 3 or 4, all distances.

Total Bets 4186
Total Amount Bet 8372.00
Wins 1014
Pct. 24.22%
Amount Won 8145.90
Profit/Loss -226.10
Pct Profit Loss -2.70%
Avg. Payout 8.03


ALL_MSA Misc Query ROI Report
nESPD >=9, dist <8F, rTRN 2,3,4.

Total Bets 2403
Total Amount Bet 4806.00
Wins 592
Pct. 24.64%
Amount Won 4284.00
Profit/Loss -522.00
Pct Profit Loss -10.86%
Avg. Payout 7.24


ALL_MSA Misc Query ROI Report
nESPD >=9, dist >=8F, rTRN 2,3,4.

Total Bets 1783
Total Amount Bet 3566.00
Wins 422
Pct. 23.67%
Amount Won 3861.90
Profit/Loss 295.90
Pct Profit Loss 8.30%
Avg. Payout 9.15

I should call this set of posts a "short course in why queries can/will drive you crazy". Too many things make no sense (which is of course why I love it, reminds me of life).

[Carl]

Nice $34 winner in the 3rd at DEL today, nESPD +10, ev1 and and XF1 to boot. Some time it's a while between drinks with these, and sometimes it's not.

later,
Carl (liking this horsey a lot cause it come up on two spot plays)

[George]

Great hit !! With your $20 winner yesterday it's time to take wife out to dinner.
Or do you, like me, put on a soulful look and announce to wife that it was "breakeven" today.

Question.......Do you "up" your bet when called on multiple spot plays??

[Carl]

George,

Yeah, it's always "break even" around here. Hehe.

I have been uping my bet (betting each spot play individually as it were) and it has been working out real well for me the last couple of weeks. I hit a double play yesterday with the SUF r7 h8, and on Friday a triple at River that was in the $20 range.

I only get maybe one in eight plays hitting multiple, so it is kind of a treat for me when one wins.

The first thread I did here in the access section was my tracking (backfitting) of same, do multiple plays hit more often? Nice for me, the benefits have carried forward.

Of course, as a few other posters have pointed out, the slope downward is fast also.

Have you checked out what happens for you when two or more plays come together on the same horse? Have you found it to be a better play? Three of my plays have turned out to be "confirmation dependent", which is to say their ROI is high (over 100%), but only when the horse they choose is also picked by another play .

later,
Carl

[hurrikane]

Nice pop Carl. I had him on 2 spot plays too. makes for a nice way to start the day. Since then mostly treading water.

About your query above. I was thinking about the rtrn>4, wondering if it wasn't the multiple hits causing the aberation you discussed. I went through my db (not nearly as big as yours) and ran the same query but with rtrn = 4 and then 5 and then 6 etc. On avg I didn't find much diff between routes sprints and even trainer rank. Was closer to a break even ROI though which could be my smaller data or the power of ESD.
The route/sprint difference could be a couple of things...difference in size of the field in routes and sprints which would change considerably the number of multiple trns in the same >4 query. And/or the number of races themselves assuming many more sprints which would cause many more multiple hits >4.
The one factor I have found with ESD and trainers that is very effective is the relationship to layoff. If you get a chance can you run ESD >=5 lay >45 trn=1or2or3
I ran each trn rating individually and I found a loss started at rtrn 4. Maybe you could confirm for me. Seemed rtrn >4 lay>45 was almost a throwout which is good. I've used it on false favorites and to back off some ESD plays that ran out of the money.

Again...congrats on anouther "break even' day :)

[Carl]

Hurrikane,

"Treading water"? I got to learn how to do that. Cashed my first ticket today on the DEL horsey, been sinking ever since. (That's why I was glad it was a double bet, don't really need to cash another, though it would be nice.)

Looks like I can confirm what you say about a "good" trainer is a much better bet with a layoff horse under this good nESP condition.

I ran four queries, I keep the number at ">=9 nESPD" for comparison purposes.

1) First query is all horses >=9 nESPD, >=45 days off. Did this one to establish a base number.

ALL_MSA Misc Query ROI Report
nESPD >=9, nLAY >=45.

Total Bets 1322
Total Amount Bet 2644.00
Wins 270
Pct. 20.42%
Amount Won 2325.00
Profit/Loss -319.00
Pct Profit Loss -12.07%
Avg. Payout 8.61

2) Second query is same criteria, except rTRN <=3 added.

ALL_MSA Misc Query ROI Report
nESPD >=9, nLay >=45, rTRN <=3.

Total Bets 631
Total Amount Bet 1262.00
Wins 170
Pct. 26.94%
Amount Won 1266.80
Profit/Loss 4.80
Pct Profit Loss 0.38%
Avg. Payout 7.45

3) Third query is same criteria, with trainer 4 or 5 or 6.

ALL_MSA Misc Query ROI Report
nESPD >=9, nLAY >=45, rTRN 4 or 5 or 6.

Total Bets 441
Total Amount Bet 882.00
Wins 74
Pct. 16.78%
Amount Won 675.60
Profit/Loss -206.40
Pct Profit Loss -23.40%
Avg. Payout 9.13

4) Fourth query is same criteria, with trainer ranked >6.

ALL_MSA Misc Query ROI Report
nESPD >=9, nLAY >=45, rTRN >=6.

Total Bets 250
Total Amount Bet 500.00
Wins 26
Pct. 10.40%
Amount Won 382.60
Profit/Loss -117.40
Pct Profit Loss -23.48%
Avg. Payout 14.72


later,
Carl

[hurrikane]

Thanks Carl....you know the only problem with treading water is you have to start swimming some time or you end up drowning. ;)

this layoff seems to be the biggest weakness in the esd number and the trainer the best way to overcome it. Also, I can't tell you have many bets I have saved not betting a low ranked trainer with more than 45 day layoff. If I can't find something else to convince me he's a throwout...almost automatic. will play in the exotics on the bottom though with some success.

[hurrikane]

Carl,
looking back over my post I probably seem like a dunderhead. I wasn;t very clear about how I did it and you were probably wondering how you could have multiple trainers with a +esd.
I pull the trainers out and them filter by esd to find a profitable range so in my data there would be multiple >4 trainers for the same race. Sometimes I forget not everyone attacks the data the same way but thanks for the query...validated my thoughts.

Too bad about the work..hope you get over it soon.