Request for Info

[Glen]

Can someone tell me the best factor(Win%) and the ROI on thier modeler for MNR. Just test all races regardless of conditions...


Thanks,

Glen

[chuck]

Here's what I had handy


07-24-01 MNR-Win Records: 279 PL: 5 Avg Par: 092 Avg Fld: 9.4
+------------------------------------------------------------------------------
¦ Factor Rnk1 Top2 Top3 Top4 Top5 Top6 LONGSHT ROI-1 ROI-2 ROI-3
¦ ----------------------------------------------------------------------------
¦ MLO 24% 46% 65% 80% 86% 93% 01/61 $0.76 $1.06 $1.16
¦ HTR 24% 42% 59% 71% 78% 86% 03/61 $0.79 $0.84 $1.19
¦ (K) 25% 43% 58% 70% 81% 88% 00/61 $0.85 $0.77 $0.91
¦ P-Scan 24% 38% 59% 72% 84% 90% 03/61 $0.97 $0.58 $1.26
¦ POW 24% 46% 58% 72% 80% 86% 08/61 $0.88 $1.04 $0.80
¦ CLA 23% 40% 54% 67% 77% 84% 03/61 $0.86 $0.78 $0.95
¦ PER 21% 38% 52% 64% 71% 76% 08/61 $0.78 $0.94 $0.97
¦ PROB% 25% 46% 65% 76% 81% 89% 04/61 $0.85 $0.98 $1.03
¦ EFF 16% 34% 44% 56% 66% 77% 11/61 $0.80 $0.98 $0.57
¦ JKY 16% 34% 46% 59% 66% 76% 14/61 $0.84 $1.11 $0.63
¦ TRN 19% 33% 43% 56% 65% 76% 13/61 $0.96 $0.82 $0.49
¦ ACL 19% 33% 44% 55% 62% 70% 09/61 $0.83 $0.85 $0.93
¦
¦ VEL 19% 33% 48% 59% 72% 79% 04/61 $0.85 $0.97 $0.83
¦ Fr1-Ev 15% 28% 40% 52% 66% 74% 07/61 $0.76 $0.73 $0.81
¦ Fr2 12% 25% 36% 49% 64% 72% 11/61 $0.61 $0.80 $0.72
¦ Fr3 11% 22% 33% 46% 57% 67% 14/61 $0.84 $0.75 $0.96
¦ A/P 18% 34% 48% 63% 72% 80% 13/61 $0.77 $0.95 $0.67
¦ E/P 16% 29% 47% 56% 65% 77% 07/61 $0.75 $0.72 $1.20 _
¦ S/P 18% 30% 42% 51% 63% 76% 16/61 $1.08 $0.85 $0.75 _
¦ F/X 15% 30% 42% 54% 66% 74% 12/61 $0.75 $0.97 $0.74 _
¦ L/P-Lv 12% 25% 37% 47% 57% 72% 17/61 $0.71 $1.10 $0.63 _
¦ (*) 16% 09/61 $0.87 _
¦ _
¦ ---------------------------------------------------------------------------- _
¦ RS Starters Won-Pla-Sho Win% Impact $ROI ITM% %Race Winners _
¦ ---------------------------------------------------------------------------- _
¦ F 0087 015 015 005 17 1.63 1.07 40 05
¦ E 0601 073 072 069 12 1.15 0.64 36 26 _
¦ P 0725 103 084 077 14 1.34 1.13 36 37 _
¦ S 0910 067 090 098 07 0.69 0.60 28 24 _
¦ R 0308 021 018 030 07 0.64 0.73 22 08 _
¦ _

[Glen]

Thanks, how did you get that thing on the bbs? Scanner?

[chuck]

I just found out the trick. On win98 on the dos shell if you click in the upper left corner on the MSdos logo - go to edit and then mark acts just like cut and paste.

[Glen]

Chuck, thank you. That is a great trick! We need to change your name to Dr. DOS!

[Rick]

Thanks Chuck.
I have Win2000 and was having trouble copying and pasting from the dos window. Its a little different in Win2000 cause you have to highlight what you want to copy first, then click the upper left corner and click Edit, Copy.

In the past I was highlighting and hitting Ctrl+C; sometimes it would copy but most of the time it wouldn't.

[Mike Moore]

According to the fine print out by Chuck, the most prolific winners are nothing less than the Morning Line published by the local track handicapper. This is free... I keep subscribing to numerous handicapping progams. Am I wrong, or are you???
Mike

[Carl]

Mike,

MLO is generally a good percentage number at most tracks. Ken mentions in one of his articles that by his count the MLO goes off as the favorite 80% of the time, and, interestingly enough, is a slightly better bet when it does not go off as the favorite (although you still lose something like 12 cents on a dollar betting it then).

IMO the best factor Chuck posts is "LP", because it shows a $1.08 profit for every dollar bet on it when it is ranked "1". This doesn't surprise me, because EVERYONE KNOWS SPEED WINS AT MNR (hence the "contrarian approach" of betting Late Pace wins money).

For comparison purposes, here is what I have for MNR MLO ranked one in my db. (Sorry Glen, I don't have any modeler statistics.) I have it winning at a slightly higher rate.
=================================
ALL_MSA Misc Query ROI Report MNR rMLO1
Total Bets----------------1758
Total Amount Bet----------3516.00
Wins----------------------485
Pct.----------------------27.59%
Amount Won----------------3001.40
Profit/Loss---------------(-514.60)
Pct Profit Loss-----------(-14.64%)
Avg. Payout---------------6.19
=================================


later,
Carl

[hurrikane]

Interesting that I have completely different findings. I've been winning money with this data..not sure why the difference. Even when I backed down to 260 races things held about the same. Anyone have any insight as to why the numbers are so different.
I'm sure you now the differences you find when you separate the wet=turf-dirt and the spr/rte data. This is all surf/all dist.


07-26-00 MNR-Win Records: 400 PL: 5 Avg Par: 092 Avg Fld: 9.4
+------------------------------------------------------------------------------+
¦ Factor Rnk1 Top2 Top3 Top4 Top5 Top6 LONGSHT ROI-1 ROI-2 ROI-3 
¦ ----------------------------------------------------------------------------
¦ MLO 28% 45% 60% 75% 82% 88% 03/89 $0.88 $0.84 $0.79 _
¦ HTR 28% 45% 60% 71% 81% 88% 08/89 $0.89 $0.89 $0.88 _
¦ (K) 28% 48% 60% 70% 80% 87% 05/89 $0.85 $1.05 $0.67 _
¦ P-Scan 24% 42% 56% 67% 78% 87% 05/89 $0.80 $0.78 $0.73 _
¦ POW 25% 44% 60% 70% 80% 86% 07/89 $0.78 $0.92 $1.01 _
¦ CLA 24% 40% 54% 67% 78% 85% 10/89 $0.83 $0.89 $1.05 _
¦ PER 25% 44% 55% 65% 75% 83% 12/89 $0.91 $1.04 $0.67 _
¦ PROB% 29% 47% 62% 72% 81% 87% 02/89 $0.90 $0.81 $0.98 _
¦ EFF 23% 39% 49% 58% 67% 76% 17/89 $1.06 $0.94 $0.65 _
¦ JKY 17% 31% 46% 59% 69% 81% 19/89 $0.87 $0.98 $0.89 _
¦ TRN 22% 40% 52% 64% 74% 81% 27/89 $1.22 $1.43 $0.62 _
¦ ACL 17% 28% 42% 54% 62% 73% 12/89 $1.03 $0.54 $0.86 _
¦ _
¦ VEL 23% 37% 49% 60% 69% 77% 02/89 $1.05 $0.91 $0.70 _
¦ Fr1-Ev 14% 30% 42% 54% 65% 76% 22/89 $0.84 $1.03 $0.99 _
¦ Fr2 14% 31% 41% 51% 64% 74% 16/89 $0.80 $0.88 $0.59 _
¦ Fr3 09% 19% 30% 40% 57% 67% 12/89 $0.45 $0.96 $0.66 _
¦ A/P 24% 39% 50% 62% 73% 81% 18/89 $1.06 $0.96 $0.60 
¦ E/P 19% 34% 50% 59% 68% 79% 23/89 $1.11 $1.00 $1.06 _
¦ S/P 14% 30% 39% 52% 65% 76% 13/89 $0.71 $0.86 $0.78 _
¦ F/X 16% 30% 42% 55% 68% 79% 12/89 $0.73 $0.78 $0.81 _
¦ L/P-Lv 09% 24% 38% 49% 60% 72% 13/89 $0.53 $0.95 $0.95 _
¦ (*) 18% 23/89 $1.03 _
¦ _
¦ ---------------------------------------------------------------------------- _
¦ RS Starters Won-Pla-Sho Win% Impact $ROI ITM% %Race Winners _
¦ ---------------------------------------------------------------------------- _
¦ F 0136 030 013 013 22 2.06 1.36 41 08
¦ E 0853 118 103 085 14 1.29 0.86 36 29 _
¦ P 1081 120 101 122 11 1.04 0.84 32 30 _
¦ S 1247 112 148 135 09 0.84 0.89 32 28 _
¦ R 0424 020 035 045 05 0.44 0.53 24 05 _

[Glen]

Mike, the modeler results for MNR are very ugly. ie nothing truely stands out except for perhaps last speed longshots that win at a very low percentage. While it is true that at many tracks the MLO is very good, it is good to break it down further to find out what types of races produce good MLO winners. I wish my moder was still available(lap top died over the weekend) so I could give you better examples, but I will use the following from memory.

If you look back at the past LS meet. MLO Favorite Non Claiming Female dirt sprinters won about 50%. Where as in certain turf races at LS the MLO won about 15%. In those 15% MLO races, about 40% of the winners paid $20 or more. Having such information is valuable when trying to out play the competition in this game. As a reminder, our competition is not the horses, but our fellow betters. After all, we are after the betting publics money...Not the horses money...

[chuck]

The original modler stats I posted were june and july to date. I was checking short term trends. I think it kinda interesting that the stats varied so much. Which gets back to the point on one of my other questions - How do you know that this is a short term trend or something that will hold up. Obviosly if you played number 1 SP in my model you win money with a roi of 1.08. But if you look at hurricane's (which I assume is from a different time period) SP is a loser. How do you know!!! Which is why I've shied away from using track specific stats.

I think that with a lot of this stuff we've become great predictors of the past not necessarily predictors of the future.

On another handicapping board they were talking about trainer stats. One guy was playing trainers 1st 2nd 3rd layoff and he found some that historically won 30%. When he forward tested it he found that they only won 22%.

[Carl]

Well,

I'm with you on this one Chuck. I personally think that what we see looking at when we look at individual tracks is "realization of the norm" waiting to happen. If I see F & M sprinters winning 50% at a certain distance and catagory where they historically win 25% across North America, I would be inclined to got the "bet against" before the "bet on", hoping the track heads down to the 25% level with the rest.

However, Glen and Hurrikane both make money with the "bet on" short term factors. And I know from his Newsletters Ken is a big fan of betting short term trends based on distance and surface.

Of course, my "belief system" goes further left from here in that I believe all you can say about this is that all "North American tracks are Universally Speed favoring" (Mostly due to the turns). I think all individual track statistics are essentially meaningless. Needless to say, I don't use the modeler, I don't believe in value of the information it gives out.

And differences in handicapping and betting strategy are what make horseracing fun. Somethings work one time, other things work another. Who would have it any other way? I bet my $2, you bet yours, and the state takes their 40 cents, you or I get back $1.60. Hehe. (Think I want the State's part of this action.)

later,
Carl

[hurrikane]

Carl, I appreciate what you are saying and think there is some value in it without question. I am curious why you devalue the modeler so much when it is doing exactly the same thing you do with your database when looking for spot plays. You see what is working, showing a profit, and what isn't. Only difference is you can be much more specific and use many different combinations with the database.

I will ask you to amend your statment that ALL north american tracks favor speed. I have picked up some very nice pays at Del from horses coming from HAW and a couple of other tracks where speed is not an advantage. These horses came off at great prices and were able to carry the front to the end at Del where speed is the play.
So if you change your statement to MOST tracks we will be in agreement....

As for the difference. Curious if the data you posted, Chuck, includes turf races? They have started running them at MNR over the last couple of months and these are heavily favored in F3 AP SP and LV. There have been some pretty high payoffs too. May be skewing the data.
I also think shippers are having a big effect..these horses may be listed as Lv1s but are much better horses and can overcome any bias the regular pool of ponies may run with.

However, my best buy at MNR in July was a ev1 esd+11 on the turf. Could be the contrarian theory or, more likely, this horse was much the best and went w2w.

PS...I say we start our own state..open a track..go with 10% take and rake it in! A few slots of course to pay the bills.

[Carl]

Hurrikane,

I argree that as always all "All" type statements are stupid (even/perhaps especially so when I make them.). But being a pighead, I continue to treat "all" tracks the same. As for HAW, I though speed held up nicely there this year's spring meet, although it stunk at last year's fall meet. I caught a few good one's there in May. For example, here is HAW runners with an * (projected to take the lead) this year from my db:

ALL_MSA Misc Query ROI Report HAW 2001 rNLDR1
Total Bets-----------------198
Total Amount Bet-----------396.00
Wins-----------------------36
Pct.-----------------------18.18%
Amount Won-----------------380.60
Profit/Loss----------------(-15.40)
Pct Profit Loss------------(-3.89%)
Avg. Payout----------------10.57


This a much better ROI than I have with "*" for the entire track population. IF I would have skipped the track based on my previous ev1 type stats, I would have missed some good bets.

And I keep seeing the "HAW phenomena" over and over. What works one meet, don't work the next, what works one week don't work the next. It was driving me CRAZY, so I decided to ignore individual track differences completely (even if they are "real"), and have been a happier capper ever since.

As to the differences between the modeler and db, to me huge, at least in the way I am using them. The short of it is that:

I use the db to look for combinations of universally underbet factors .

The modeler only lets you track single factors, not combinations. I like to mix up a soup of "some good, some bad" to get a profitable historical play. The modeler don't do that at all.

later,
Carl

[MikeDee]

Carl and I disagree on if different factors have different results per track. I am of the opinion that they make a difference and are not just regressing to the norm. I wasn't going to commet on this but but since Gordon Pine sent a newsletter on the subject I just couldn't resist. In his newlsetter he wrote.



At Colonial Downs, early pace is important. The horse that figures to be in front at the second call wins 28% more than it should given its odds (has an A/E of 1.28), and generates a 1.09 ROI (9 cents profit for each dollar bet). (These stats are derived from a handicapping program I designed.) It’s also a track favorable to longshots: Over-achievers have an A/E of 1.20 and an ROI of 1.02. Pretty good predictors. But take these same predictors to another meet: at Arlington Park, the top second call horse has a .81 A/E and a .69 ROI. Over-achievers have a .92 A/E and a .79 ROI. So, at Arlington, one predictor earns 40 cents less per dollar and the other earns 23 cents less per dollar than at Colonial.

Gordon states my point better then I can, since I copied from his newsletter the least I can do his give him a plug. If you would like to get on his mailing list vist his site at:
http://www.netcapper.com

[hurrikane]

Carl, pighead isn't really neccessary. :D
So many things work so many different ways in this game, the only important thing is that we win and have fun doing it.
I agree with you on the HAW info..I had a great time at HAW...but I was playing Vel..and ap1 sp1 ep 1...this I found in the modeler and mixed and matched. When the meet was over it turned out ev1 and esd had done real well. However...I still find great overlays at DEL and ELP. These have always looked like horses that died in the stretch at HAW and then they carry all the way for a great price. I know speed was there looking back at the data..or maybe it was just a coincedence that the speed was on top.

I understand exactly what you are saying about the db vs the modeler.
I suppose my question is 'take some good points..take some bad points'....if everything works back to the norm then what are the good and bad points..

[Carl]

Mike,

I am sure Gordon is right about what has happened at Colonial and Arlington.

Of course, what we all want to know is what will happen at Colonial and Arlington.

Will the "trends" hold up? Indeed, are they even trends? Ron Tiller makes a great point at Netcapper IMO when he points out
=================================================
"Taking a random 50% series, like a series of coin tosses, short term sequences within a long series will be very streaky. Many statistics teachers, as a demonstration for new students, ask them to write a series of 100 random H's or T's. The teacher is always able to tell actual random coin tosses from the student's "random" tosses because students always have more alternating H's and T's than actual random coin tosses. That is, students will have a lot of sequences like HTHTHT. In actual coin tosses, there will be lots of clumping, with 5 or 6 heads or tails in a row. Few people will include such streaks in their answers, enabling teachers to mystically tell the real from the fake series."
==================================================

Anyway, we shall see.

BTW, my cynical attitude towards "models" of particular track tendancies is mostly cause by Florida turf. For the past two years, I have found both the TAM and GP surfaces playing strongly to speed for a month or two at a time, then by the time I was on it, nada. Sure, it could be rain/lack of, stupid jockey colony, tides or ?? However, for the time being, I am going with "random coin toss streaks" for the phenomena, and I am going to vote for AP ep1 non-dominance and CNL ep1 dominance as being "random coin toss streaks" as well.

I'll run AP ep1 and CNL ep1 later on today for the first part of July (till last Sunday), and I'll re-run this Monday for this week including next Sunday. And I'll bet my July funny dollars they reversed. (Maybe a good thing it's only funny dollars hehe. But you never know......)

Hurrikane,

Ken's article a month or two back about "streaks" was cool here also IMO. And I do think some percentages hold up over time, it is just that except for crazy FG, I haven't found a track that I think is anything but "speed favoring" yet over the long haul. And yes I will bet the early speed at FG again and probably get pounded again (although ep was very good there the first two weeks last year as I recall, so I better "hit a few bomb early then...." Hehe.).

later,
Carl