The Impact of Wagering Earlybird vs. Live

[dvlander]

I want to ask the brilliant handicapping minds on this board a theoretical question.

Suppose I have a 10,000 research race sample of a small group of Win only spot plays that are all positive ROI to different degrees. In this large sample, the overall win percentage is 16% and the average win mutuel is $14.00 for a resulting ROI of 1.12.

Obviously, all of this test work was done with completed races after scratches, surface changes etc. If we assume (believe me, I know that is risky in horse racing) that results going forward will be similar, what kind of negative ROI variance can we expect if all betting is done earlybird with conditional wagering without those day-of-race changes? My schedule does not allow me to hang around the ADW all day waiting for a horse to scratch.

Before you answer, the impact of betting earlybird is not always negative. I had a selection a couple of weeks ago that came in at long odds and it would have been eliminated if I had to consider post-time changes. With a pre-scratch long-term ROI of 1.12, you can give back a few basis points but not much more than that.

I appreciate the responses.

Dale

[andymays]

Scratches and surface changes can dramatically change the dynamics of a race.

It's never good to bet early unless you have to!

[markgoldie]

Having done the research, you are probably in a better position than any of us to make a definitive comment.

If the handicapping processes are dependent on track condition, jockey, and pace match-ups, then the answer is probably that scratches, jockey changes, and track-condition changes may have a substantial impact.

On the other hand, if the system(s) are based more on current form, class, trainer, figs, etc., chances are better that the results will hold. Remember, many race-day changes that appear negative also raise the odds on the horse as those engaged in live wagering adjust their opinions. Therefore, you might expect a "wash" with certain types of changes that are not part of your essential handicapping theory.

[BlueShoe]

A major problem of betting very early is the late scratch that we cannot anticipate and adjust for.A single scratch can completely alter the pace scenario.It can often kill what we thought was going to be a strong play,or sometimes,it will create a play in a race we were going to pass.If the scratch happens to be the heavy M/L favorite,that 5-1 shot that we were expecting may wind up as the 8-5 favorite,and no bargain.Then there is weather;a change in track conditions,and even worse,a turf race rained off and switched to the main,with the usual multiple scratches.Good general advice would be not to wager hours before the race if it can be avoided.

[formula_2002]

Based on my own results, I was encouraged to restructure my data base and programs to study this. Here is a bit of it for about 20,000 horses

Horses >0 to <2-1 in races where no horse was scratched.
Average odds, 1.32-1
Roi =.80

Where at least 1 horse was scratched
Average odds, 1.289
Roi=.827

[misscashalot]

Scratches and surface changes can dramatically change the dynamics of a race.

It's never good to bet early unless you have to!

never say never

[andymays]

never say never


This really is one of those no brainers. If you have a choice never bet early!


Why would you if you don't have to?

[fmolf]

This really is one of those no brainers. If you have a choice never bet early!


Why would you if you don't have to?
If you have to bet early get the weather report and try to look for the best speed horses.These front runners will be less affected by late scratches..they will even be aided in certain instances where other speed scratches.this will drive down the odds but you said you are betting conditionally correct?

[Handicap Analysis]

I usually don't post in these forums I just enjoy reading, but you ask a question that has been a long time study of mine.

My original theory was oven the long term your results should end up around the same, but in practice it hasn't worked that way.

Like you I have a large database of the last 5 years.....well over 50,000 races, and I've developed some systems that produce great results going forward.

While the statement "betting early is never correct" is true I don't agree with it. Most people cannot spend all day waiting for changes to rehandicap. Most have work and families...but still would like to take advantage of the inefecencies of horse racing.
Prebetting, as I call it is a good solution. But I have found in studying the difference that the results are nowhere near the same. I have been keeping track of my picks before and after scratches for the last 2 years....and the after scratches results are almost always better.
I even had a month where scratches eliminated a $50 horse, but still out performed prebetting. I can send you my results if you like.

My prebetting results are still profitable, but usually only half as much as after scratches.

[dvlander]

My prebetting results are still profitable, but usually only half as much as after scratches.


Handicap Analysis:

I really appreciate you chiming in on this because you appear to have the research basis that I was seeking. So when you say that you only achieve half the profit betting early, may I reasonably expect that my long-term 1.12 ROI in 10,000 race testing will approximately be reduced to 1.06 ROI if spot play performance remains similar?

Again, I really appreciate your insight on this because I know you had to do a ton of research to find out the answer.

Sincerely, Dale

[Handicap Analysis]

may I reasonably expect that my long-term 1.12 ROI in 10,000 race testing will approximately be reduced to 1.06 ROI if spot play performance remains similar?

I would say it will probably be a little more than 50%. I would expect anywhere between 50-70% reduction in ROI each month. That is usually what I see, with the occasional exception when prebetting out perfroms after scratches. Once again I have 2 years of specifically monitoring this.

Im a little concerned with your ROI, at 12% it is probably to low to prebet your specific system, unless you understand your prebetting will return a little over break even and you are receiving rebate to add 3-5% in returns.

Like I previously said, I always thought that it wouldn't make much difference, but I have been surprised in the reduction of profits while prebetting.

Good Luck.

[DeanT]

Fascinating discussion. I hope you two keep it going.

DV, can you invest in a netbook and do scratches at lunch time? It sounds like it might be a good thing.

I wonder: What if you just asked for higher odds on all your plays via conditional wagering. Would that not preserve your bankroll a little more and protect it from late changes a bit?

[DanG]

It’s certainly not exclusive to them, but to some software capper’s scratches can be their best friend. The ability to instantly re-calculate the field, re-sort etc can be invaluable over some of the competition.

Scratches are generally a pain in the…, but with auto scratch ability (HDW / HTR) it can more then compensate for the reduced field size with dynamic ratings.

I remember my old DRF / Ragozin days and on heavy scratch days my desk looked like Times Square the morning after. :eek: Red marks everywhere and the always fun routine of re-sorting post positions, making sure of horses drawing in that were on another page etc.

Between the new auto scratch ability and the various ADW’s offering conditional wagering the reasons for pre-betting completely cold have been reduced. If you can’t adjust for various reasons…you just can’t and so be it, but I have a hard time believing knowing most changes & ballpark prices wouldn’t move your game up regardless of your approach.

[dvlander]

What if you just asked for higher odds on all your plays via conditional wagering? Dean, that is a suggestion I have considered but I thought I'd try to get a feel for the ROI impact of the earlybird betting before making a firm decision. Even if reality sets in and the 1.12 ROI becomes something that is barely above breakeven pre-rebate as Handicap Analysis has indicated may occur, does it not make some sense to continue the churn and at least take home the rebate? I think requiring higher odds would undoubtedly improve the ROI but that strategy reduces the number of plays quickly and perhaps the total profit dollars would be less than leaving things as they are.

Based on all of these very worthwhile responses (which I appreciate immensely), I may take another look at the eight spot plays themselves. Maybe I can tweak and squeeze a bit more ROI out of some or maybe eliminate altogether one or two of the more marginal performers.

I'd love to find a way to play post-scratch but my schedule simply will not allow it anytime soon. It's either bet earlybird or don't play at all at this point.

Great guidance folks.

Dale

[ryesteve]

When you do your pre-scratch selections, do you "scratch" AEs and MTOs? As long as you check the weather ahead of time, I wonder if this might help a bit.

[Jeff P]

Dale,

You're not alone. I actually know several players who, because of time/scheduling constraints, have no choice other than to play early bird before scratches are announced. Formula2002's results showing higher roi pre scratches vs post scratches are interesting. True, his analysis involves a very specific set of horses. But the results are interesting nonetheless. He and I have had several phone conversations about this. One of the things I'm currently working on as a program update is the ability to give a JCapper user two options when building a database:

1. Remove scratched horses from consideration before number crunching is performed.

2. Include scratched horses when number crunching is performed.

Of course option #1 is the current info model for reasons already stated in this thread. But for those who, because of time/scheduling constraints, have no choice other than to wager before scratches are announced - option #2 would allow the player to perform R&D and develop models based on the environment in which he or she must play.


-jp

.

[dvlander]

Ryesteve: I had not considered that. It's a worthwhile suggestion. Thx, Dale

[dvlander]

You're not alone. I actually know several players who, because of time/scheduling constraints, have no choice other than to play early bird before scratches are announced. Formula2002's results showing higher roi pre scratches vs post scratches are interesting. True, his analysis involves a very specific set of horses. But the results are interesting nonetheless. He and I have had several phone conversations about this. One of the things I'm currently working on as a program update is the ability to give a JCapper user two options when building a database:

1. Remove scratched horses from consideration before number crunching is performed.

2. Include scratched horses when number crunching is performed.

Of course option #1 is the current info model for reasons already stated in this thread. But for those who, because of time/scheduling constraints, have no choice other than to wager before scratches are announced - option #2 would allow the player to perform R&D and develop models based on the environment in which he or she must play.


-jp
Your post is exactly why I am an enthusiastic JCapper user and supporter. That upgrade would be a godsend to someone like myself. Excellent news!

Dale

[Handicap Analysis]

does it not make some sense to continue the churn and at least take home the rebate? I think requiring higher odds would undoubtedly improve the ROI but that strategy reduces the number of plays quickly and perhaps the total profit dollars would be less than leaving things as they are.

With even a 16% win rate and 12% ROI you’re losing streaks and profit swings are going to be huge. You will have months where you will lose over $100 on a $2 bet. As long as you understand this and allow it to work several months before worrying about results you could continue with your current system. Most people cannot handle this type of losing streak without having it affect their emotions and betting decisions.

Also, you are correct about requiring higher odds, you will increase your overall ROI but you will also reduce your profit.

[andymays]

Combined with the late scratch problem paying $8 for parking is another reason not to bet early at Del Mar!

All you can do is laugh! :D

This is one of the things they go over on the Roger Stein show today 7-26!

Live 8:00 am PST @ http://www.rogerstein.com/

Archived Show up an hour after the show @ http://www.rogerstein.com/radio/archive2.asp