Trifecta System

[foxwood]

Hi Folks,
I'm only a recent recruit but I've had a lot of pleasure from reading the threads on these forums and would like to make a contribution.
I have an idea for a system for trifecta betting which I would like to test out over several months of Saturday racing. The trouble is I have to enter all the data for sometimes 16 - 18 horses for 6 or 7 races per meeting for four meetings per weekend all in longhand as it were into Excel. After entering just a few weeks of data I'm starting to go cross eyed. Does anybody out there have an easier way to get this data from a database to Excel to test my theories? Obviously if every member of the forum were to have access to the system(should it prove to be any good) the tri divi would disappear through the floor, so I would rather restrict distribution of my idea for this system to just one or two, preferably one.
I would like to stress that this system was not created by fitting results into a pattern and then trying to repeat them as most of my previous inglorious failures have been. Rather, I had an idea, created a system based on that idea and tested it.
So far in the 8 weeks that I've looked back at it has found:
13/11/04 Bris R3 $14732 for $60,
18/09/04 Syd R8 $11147 for $60
plus several smaller divs ranging from a few dollars to a a couple of grand but to be honest, an awful lot of races where I didn't even get a placed horse.

If anyone can help I'd be pleased to hear.
Cheers

[Duritz]

I run my own database and query it in excel, returning all manner of detail about every horse, meaning of course that all I have to do is input the fields. I am sure I could help you. Let me know if you'd like to give it a crack.

Duritz.

[crash]

Hi Foxwood,

I'd be interested to know the reason you are backing so many races ? Without wanting to know your system rules, Is it a numbers system ?

Even with the Tri. there is seldom more than 3 or 4 races [often less] that make sense to bet on at any meeting.

Unless you are laying horses on Betfair for very small % wins, to make a profit, quality of the selection process generally thrashes the quantity method.

[laish]

Hey FoxWood,
I can certainly help if you ned it. I run a program for my systems so I just enter in the criteria and it pulls up the selections for me, and I have data running for roughly 4 years, I can then easily enter them into Excel with additional stats if you like. If you want me to give it a go please contact me at laish at optusnet dot com dot au. Also if it helps I have my own system for trifectas that I use so therefore I would not be looking at useing yours, if this helps to decide. It has pulled up 100% profit above turnover and 470% on First Fours.

[foxwood]

Thanks for the replies fellas.

LAISH, I'll e-mail you within the next few hours with some details and take you up on your offer.

DURITZ, Ditto the above if you post an email address.

CRASH, You are probably right about looking at too many races. From what I have done so far it seems that I should concentrate on large(ish) fields and no 2 or 3 y.o. races.

Good luck this afternoon to anyone having a try.

[Duritz]

Send to me at kingsandmadmen with after that the @ symbol of course and then hotmail dot com. Would be interested to see how I can help, and of course perhaps improve upon.

Duritz.

[haveadip]

A question the people in here who have stored Excel results, which must be very time consuming. Is that do you people make a profit? If so how great and do you think its worth all the effort? Also how many people in here who have there own mathematical system add form analysis into the equation to increase winnings?

[Sportz]

Also how many people in here who have there own mathematical system add form analysis into the equation to increase winnings?

Good question. This is one of the difficult things about system betting. If you use systems, often they come up with selections that you don't honestly think can win or in races where you think another horse has superior form. As a result, I always mix systems AND form analysis. A good example was Caulfield R7 yesterday. The new system I'm trialling picked Trick Or Treating, but in my own mind, Hollow Bullet was basically a good thing. So I backed them both, having more on Hollow Bullet because of the shorter price. Then again, I am used to backing more than 1 horse per race. I do that 90% of the time. For people that like to back just 1 horse per race, I'm not sure what they do.

[Sportz]

Haveadip,

I've just read your question again and I'm not sure I answered you properly there.

[haveadip]

Na thats the response i was after for the third question, thanks. Just interested to see how many people have purely mathamatical systems that are succesful.

[foxwood]

haveadip,
I only ever use a mechanical system on the races, never form analysis. I just don't seem to be able to get the hang of it. There must be something I just can't see. Even when I look at the form for a race after the event, the winner is never obvious to me. So I put my faith in trying to find a magic formula and the older I get the further away that seems too.
However, the search for the system rather than the betting, (which I usually don't do anyway) or the race watching (which I never do), is the continuing fascination for me. I spend hours devising systems and then testing them out against old Winning Posts. Finding the repeated pattern is the thrill. Unfortunately, every system seems to work for a short while then peter out (and I've seen that complaint aired by others on this same forum.) When I do have a dip on a system selection I would be petrified of trying to second guess it by applying my paltry form analysis techniques and I know what would happen if I did.
Regards

[crash]

I think similar to Sports on the matter of adding form analysis to a system but instead of backing a horse in a race that looks obviously more qualified to win than the system's selection, I just leave the race alone as I like to keep my system bets separate from my form bets. I have been 'doing the form' for over 30yrs., so I don't have a confidence problem in this area.

I am amazed that many punters will spend huge amount of time working out systems yet won't put in a few yards to learn and practise one of the basics of punting ...form study.
There is reams of info. available on coming to grips with race analysis and many different approaches to the subject from the more complex 'formline' style commonly heard on the radio, to the simpler approach of giving each horse in a race points [or to start with, even the first 5 in the pre-post betting line] for meeting certain criteria. Both methods are valid and the more complex is in no way superior or more successful to methods that are far easier, but just as valid.

[Duritz]

I don't actually think a mechanical system can work long term. I watched my old man come up with mechanical systems ad nauseum growing up, and I remember each time him saying that he thought this one was going to win blah blah but they failed when tested with money. He has a first class honours degree in economics from Melbourne University. His father, an owner, breeder and sometime SP bookie, kept piles of the Sportsman newspapers in his study and was always researching systems. He never came up with a winning one either.

All the men (and women), and all the millions of hours of study that they have put into it, and do you actually know of a winning system? It's like cracking an impossible code, and I have a small theory as to why:

Think of humans. Think of all the different ones you have known or known of, all the weird, colourful, positive, negative things they have done.

Now try to make a system filled with mechanical, arbirtrary rules to explain them.

Let's give it a crack:

1 - When on a train, they will sit down. Or stand up.
2 - Men like women. Women like men. Some men like men. Some women like women.
3 - If one human attacks another, the other will always fight back. Sometimes he will not.
4 - Humans like meat. Some don't.

OK I could go on, but I don't think I'm making my point correctly - you can't explain humans. You can't set up rules which classify them. We are a complex, beautiful, ugly, boring, funny, smart, ignorant, proud, arrogant, humble, generous, selfish race, infantile in our evolution, completely obsessed with ourselves.

When all is said and done, horses are just as complex as us. Hell, they can drown in a puddle of water for God sake. Arbitrary rules fall too far short of explaining what they can and can't do.

The other thing is this: The human mind is a very complex thing. Einstein alluded to his belief that the best and brightest of us (ie him) only use up to 10 or 15% of it. It is capable of immense understanding and powerful insight in a fraction of a second, and that is without us even concentrating. In that respect, the market on a horse race is generally pretty right, because the combined brainpower of all the people out there doing the form pretty much mean that the right horses go out the right prices. (A good example of the combined brainpower thing was Sportz's ratings comp the other day - the combined ratings of everyone had both winners in the top two). No pre-ordained rules containing sweeping statements about horses in certain situations can be more accurate than that combined brainpower which sets the market. That's why - in my opinion only of course - systems pretty much have to fail.

There is a way to beat that combined brainpower, and that is what Crash alluded to: you have to put in the time to do the form. Basically, put in more time to do the form than what the average of the combined brainpower pool did, and do the form properly. Learn how to do the form if you don't know, from books like Don Scott and Andy Beyer if you like time analysis, but ultimately the best way is to learn from your mistakes. It's no quick answer, it's a long grind to understanding the best way, and even when you're like Crash and are confident in your form analysis, it doesn't mean you'll do it the right way every time. You'll still get heaps of horses wrong, but that's OK. If looking back after the event the reason the winner won isn't apparent then you need to look harder. Then, when you're in the enviable Crash position of being confident in your analysis, the other half of your job begins: how to punt.

I never used to believe them when they said that form was only half the battle, that punting properly was the other half, but it's true. I know that now, it's been a costly lesson but I've learn't it (I hope). I still go out and have some very muggy bets sometimes, and they lose, but most of the time I stick to what I know is the right way to punt, and to answer haveadip's other question, his original one - people can win on it, but if doing the form etc is a grind then you probably aren't going to stay the trip. You have to love it to do it, otherwise you'll burn out.

Duritz.

[crash]

Doing the form and betting on 6 [good] races on Saturday instead of 30 might help !!!

No punter who wants non-stop action can come out in front ...nobody.

Those that need to bet on every race needs a shrink, not a winner.

[foxwood]

Crash,
If you are offering advice to me in particular re the number of races to study and bet per week it's unnecessary as well as offensive in tone. As I pointed out in the original posting, I have been entering that many races per week simply to test my theories and I rarely bet anyway. If, on the other hand, your comment is advice for the world at large I couldn't agree more.
And who the hell is Sinartra?

[crash]

Foxwood,

Yes, on seeing my post again I guess you might believe it was directed at you [I don't think you had previously said anywhere how often you bet though]. Sorry about that impression but I was thinking out loud about so many punters I've often seen at the track betting on everything that moves after a 2 sec. look at the form guide.

[Merriguy]

Great post, Duritz. I hope that most forumites take the time to reflect on all that you have said there in a very long post. A mighty lot of wisdom!!

I particularly liked the words of experience based on your father's and grandfather's endeavours. Unfortunately, most(?) of us I am afraid think that if we could only 'crack' some system it would be the end of our problems. Just pie in the sky.

Of course, this is not to say that some can just see this as an enjoyable way to spend some leisure time. More power to them :). But for those who hope to make it the basis of some future life style, there is the need, as they say, to 'put in the hard yards'.

[foxwood]

Good Morning Fellas,

I too appreciated your post Duritz. There were some interesting points in it but I'm afraid there were some red herrings too. As far as I can see you are contrasting form analysis with trying to create an algorithm to explain the entirety of a human being's activities and proclivities. This is obviously not possible, not even if you apply it to horses which incidentally, are not as complex as humans (and all the better for it in my opinion). The idea behind trying to create a system I reckon is to try to predict the outcome of a very limited period of time in a horse's Saturday afternoon under very strict and pre ordained conditions. i.e. He's running from A to B with a midget on board whacking his rump, carrying a known impost against several other all in the same boat.
I also have another question; where does form analysis end and a system take over (or vice versa). It wasn't in your post I know but somebody suggested form analysis could be done by allocating points for certain features of the horse's form or his position on a pre-post market. Does that amount to form anaysis or is it using a system? If it's the former then I'm a form analyst as well as a system compiler.
Cheers fellas and good luck for this afternoon.

[Duritz]

Yeah I don't think I was really getting my point across with that making a system for humans stuff. I read it at the time and got a bit confused myself.... And you're right - horses aren't as complex as us, but they're probably more unpredictable than us because at least we have the ability to think rationally. (though often not at the track) Without getting into detail as it's 11.14 am on Sat and I am just starting the form for today (!) I think form analysis begins as soon as you have to make a decision on a horse, like do I think it can get the trip etc. There are no doubt levels of form analysis, and of course people mix analysis and rules. One that comes to mind is anyone who looks at the top five in the market and does the form solely on them. That's a system and form analysis combined. What I don't think can win long term is a purely mechanical system, no input from the human. My old man hates having to make a decision because he's always fearful of "pulling the wrong rein". In that respect he prefers systems.

(Ironically, lately he just bets on my ratings which is for him I suppose a system, which therefore means I must be convinced it can't win lol!)

Anyway, we'll no doubt chat more about this.

Good luck today.

Duritz.

[woof43]

When I hear people talk about being picky about which races they choose to play, I always have to smile. What it says is, they have a "system." That system doesn't work very often. So they study really hard to find exactly what kind of race the system DOES work in. So now they get to play a handful of races with their method, and hopefully they'll now make a profit. Reminds me very much of the old saw, "even a broken clock is right twice a day" don't you think? Granted, these people are being smart in reducing their play to their chosen races. After all, that's the only logical way to make a profit using that particular handicapping method. If you can't improve the method, you MUST eliminate races where it doesn't work.

But the crowd plays EVERY race. And they make the same mistakes in EVERY race. You can spend your time with a crippled handicapping method, praying for a race that matches it, or you can learn how and why the crowd wagers the way they do, and start playing a very high percentage of races on every day. All it takes is to let go of one's habits and traditional notions, and start accepting how the world really works!

[Duritz]

Perhaps.. I don't think there's anything wrong with not betting in a race full of horses who have never raced before, and I certainly do think there are races which are easier to analyse because of less things left to chance, like horses first up from a spell (are they forward or not) etc. If you have a 1600m race and all the horses have had at least a couple of runs in etc, this is a better betting proposition than the blue diamond preview.

Duritz.

[foxwood]

There are systems and there are systems. I think most of us on this forum are not countenancing the mindless things such as "horse names with an x in them or horses with four white socks." (Hey, could one of you database guys check those two out for me?)
The kind of approach to selecting (and possibly even backing) horses that I have in mind when I use the word "system" is in fact just an automation of a logical selection method. That selection method would probably only consist of the same criteria that any form analyst would use.

[crash]

There seems to be confusion about the difference between a mechanical system and form analysis. With the later, personal judgement is involved and with the former it isn't.

Being discretionary with the type of races to bet on is 'how the world works' for anyone serious about making a profit at punting. Welters would be at top end of races to bet on and maidens at the bottom end. Giving all of them an equal chance of making a profit from is the main reason most of the punting world looses.

[Duritz]

I agree with crash. A system is automated. As I said earlier, I think form analysis begins when you have to make a decision about the chance of a horse.

When I talk about a system, I am also not talking about "third letter R" as one lady I used to work with betted on, I am talking about mechanical, rigid rules which allow no human intervention, like can be seen in the "up up and up" thread, ie up in class up in weigght, up in distance. Form analysis is not a system because it requires imagination and thought and decision making, hence the word analysis.

Duritz.

[syllabus23]

Form analysis is not a system because it requires imagination and thought and decision making, hence the word analysis

Analysis is when a whole is broken into it's individual parts, as opposed to synthesis where each individual part is formed into a whole.Systems men break down the whole into individual parts and assign a value to each part.From the sum of those values comes a rating.In other words they analyse and then synthesise.

The value given to each part requires imagination,thought and decision making too.The really clever guys then create a computer program which enables them to take the legwork out of it.

After that comes the really intelligent part.They keep it themselves.

[Duritz]

However the initial rules programmed into that program which then takes the leg work out of it are arbitrary and inflexible. "Doing the form" is about treating each individual horse separately, an automated program doing the form treats them all the same.

[woof43]

Instead of wasting time HANDICAPPING, people should be thinking about wagering. Period. The average punter CANNOT gain a handicapping advantage over the crowd in 15 minutes by looking at the program. But he or she COULD gain a wagering advantage, if they would stop thinking about the runners, and START thinking about taking money away from other players.

In my handicapping, I spend zero REAL time doing anything. I have computers that do all the work. But when I was developing my software, I probably spent ten times as much work on wagering and understanding the crowd as I EVER did on ranking the runners and figuring actual probabilities.

I or any good player can walk blind into a race meeting with a program in my hand, look up at the Tote board, and literally write down, immediately, in order, the runners in the race from best to worst. That takes one minute, and only one minute. And the BEST HANDICAPPER IN THE WORLD isn't going to do a better job than that, at least within about a 3% level of accuracy! I, nor anyone else can handicap 100 consecutive races, and hit significantly more winners on the nose than the crowd.

So when people spend a lot of time splitting hairs, what they are doing is trying to find either A) the race where the crowd DID screw up (a rare event) or B) the race where your particular handicapping bias works better than average. Again, that's a small minority of races.

People laugh at playing nearly every race. But think about this. The crowd make EXACTLY THE SAME KINDS OF MISTAKES ON EVERY RACE, IN EVERY CARD. All you have to do is understand what those mistakes are, why they are made, when they are made, and what advantages that gives you as a wagerer.

Here are a HANDFUL of wagering inefficiencies based on trifectas that can be exploited.

1. People bet the 3-Box trifecta. This introduces two inefficiencies. First, the favorite finishes in first place far more often than in third place, so the tri-box puts too much money on the favorite in third. Second, the worst runner in the box will finish in third much more often than in first. Thus the worst runner in the box has too much money on him in first place.

2. People bet the 4-Box trifecta . Same general situation as above, but more subtle.

3. People DO NOT bet the 5-Box trifecta, but ambiguity in the middle-class races leads to the use of 4-box tris that effect all but the worst runners in the race. The key to understanding the effects of the 4-box tri (and somewhat the 3-Box tri) is in segmenting races based on the strength of the favorite, or the top two runners.

4. Once the strength of the favorite goes past a certain level, a significant percentage of the crowd will switch from boxing to Standouts with the strong favorite as the only runner on top. This usually opens up plays with a strong favorite in second place. Second-place isn't overplayed by the box bettors, and when things switch to Standouts, the money moves from second to first on the strong favorite. Strong favorites still maintain a high percentage of second-place finishes, and they become overlays under most circumstances.

5. People overbet the 1-2-3 trifecta box (by Number, not rankings), and the 1/23/23 Standout. If your wagers include those bets at all, you're throwing money away.

6. The Tipsters selections in the daily's are certain death. It only takes a handful of people playing those picks to totally hammer those wagers into underlays. Avoid the straight trifectas, and the boxes around those selections.

7. Of course, where the crowd OVERBETS (like #5 and #6 above) there are opportunities created elsewhere. If you get a situation where the Daily's Tipster picks 1-2-3 (or 2-1-3 or that box of three in general) you are GOLDEN on almost any other play. Then is a great time to go very wide and very deep, because overlays will be like shooting fish in a barrel!

And so on

I know this is out in left field compared to what most people are thinking about, but I can't tell you how important this stuff is if you actually want to approach the wagering seriously. There are places where your time is valuably spent, and there are HUGE places where your time is effectively wasted. Not one person in 200 knows that. That's why some people CAN make money. And it's not because they are any better at picking winners

[Duritz]

Thanks woof, thought provoking and interesting post.

[moeee]

Gee Woofer.You frightened me!
Made the hair on back of neck almost stand up.
Had to save that page and will read it again when I stop shaking!
Well done.

If it weren't so scary and captivating,I would have asked how many times did you time out whilst writing that epic.

[DR RON]

Two factors required to make a profit on the punt.
1. Picking winners
2 Getting value for every bet you make.

For those with a limited time to devote to those tasks they will probably concentrate on the area they think they do best at.The trouble is that spending the majority of your time on finding winners wont do you much good if your not getting value from your betting, especially if your form analysis throws up the same findings as the general betting public. Even though I enjoy doing weight ratings I find that the time they take to do thoroughly gives me less time to concentrate on getting the best value for them. I guess the bottom line comes down to the question, Are you making money from your betting over the long term?? If you are then keep going with what you are doing, whatever the method.

[Shaun]

Woof43....i would have to agree on your observations....if anyone has had a look at the trifecta pools lately they are larger than the place pools most times...with less people winning them....with more and more information out there you would have to think that more and more people are getting it right in relation to form study....can anyone give me stats on the % of winners under $10 or the % of winners from the top 5 weighted horses....this shows you that the form is already done for you....i know it is great to beable to rate a race and find errors and have those errors salute at good prices....but i would have to say it is harder to find those winners....i have been doing ratings for over ten years of some sort or another and things have changed....we have to move in another direction if we are to maintain our edge...i do belive that there is a mathamaticle solution to all this and thats where i have been heading of late....yes i know you can't back every runner in the market to win but there are other ways i am sure of that.

[woof43]

Follow this idea... the crowd is a VERY GOOD handicapper, right? Their top-ranked selection wins far more often than their longest-odds runner. But why isn't the ROI on the crowd's TOP SELECTION exactly the same as on their WORST SELECTION? Why aren't both of those numbers EXACTLY THE SAME? You would think, if handicapping is important, that those two numbers would be nearly identical. After all, the crowd should be exactly as good as saying runner X is BEST as they are at saying that runner Y is WORST. But in reality, those two ROI numbers may be WAY DIFFERENT. Why is that??? Because handicapping just isn't as important as wagering. And while the crowd is really good at handicapping (as a conglomerate) they are incredibly bad at wagering. This is an exercise in psychology, not an exercise in picking winners.

[kenchar]

woof43,

One of the best and true to the fact posts I,ve seen in a long time.

[crash]

Woof43 wrote:

...'This is an exercise in psychology, not an exercise in picking winners'.

A nice bit of insight.

[Duritz]

Ok well if all that is true re the underbetting/overbetting of given combinations, does the same apply to the win pool? IE is the win pool biased towards the tipsters picks or the top few in number order, and if so then is this where handicapping and pricing and betting overs can win? Basically, is the mug money doing the same to the win pool that you describe it doing to the trifecta pool, and can this be used to advantage?

Duritz

[woof43]

We live in a world of probabilities, not certainties.

I guarantee that if people play the "best runner" they are going to get hammered. (Assuming they play more than some tiny percentage of races.) There is NOTHING that says the best runner owes you a profit. And people that spend their precious time trying to distinguish which of the top two or three is REALLY the best are just wasting that time.

At Melb dog track the 2-1 favorite coming out of the #5 box has an average ROI of -9.02%. And, a 2-1 favorite coming out of the #8 box has an average ROI of -33.46%. So, on average, your 2-1 favorite starting from the #5 box is worth 91 cents for every dollar you play on him. Put the same dog with the same odds in #8, and now he's worth less than 67 cents per dollar wagered!
Within a percentage point or two, the 2-1 favorites in #5 and #8 will actually win just about the same number of times.
But the crowd ALWAYS OVERPLAYS THE #8 BOX AND ALWAYS UNDERPLAYS THE #5 BOX! If you do your handicap, and get the same strength of winner in #5 and #8, you SHOULD NOT PLAY HIM THE SAME WAY FROM BOTH BOXES! If you do, you're being a fool. That is why I chide those who do their handicap, and then get up and play their favorite in the same kind of wager, no matter what kind of race it is, and no matter what box their dog is in. They spend all that time finding a favorite, then they likely throw away whatever advantage they might have by making a stupid wager.
And that is what I mean by wagering being more important than handicapping. Spending your time splitting hairs between the #5 and the #8 is just plain dumb. Spending time understanding the wagering picture in a race is very smart. Understanding the wagering situation means that I don't have to look too closely at the #5/#8 situation AT ALL. If those two are close at all, take the #5!
It is SO MUCH about wagering, and not handicapping.
Take a look at yet another wagering statistic from the same track using the same 2/1 fav, If I take my SAME RANKINGS and instead make a three-dog box tri, what then?

3-dog box: 80 hits out of 1286 races (6.22%), average payoff is $89.66, and the ROI is a miserable -53.52%. (BTW, the 6.22% hits is very reasonable for a three-dog box.)

So,I take the EXACT same handicap and bet it three ways:

123/123/123 (3-dog box) (stupid, -53.52% ROI)
or
12/1234/1234 (meerly dumb, -1.07% ROI)
or
12/1234/5678 (sweet, +62.84% ROI)

The HANDICAP isn't important. The WAGERING means the world! Just LOOK at that spread. The BAD wager has a 53% LOSS and the GOOD wager has almost a 63% PROFIT! All based on the exact same handicap.

In all my years at the track, I've never seen ANYONE that can improve their HANDICAPPING from a 53% loss to a 63% profit. But you CAN improve that much if you study wagering.

[woof43]

Duritz,
This is where Handicapping, pricing (assigning probabilities) betting overs and winning starts.

The key to understanding handicapping is to realize that you can only get so good at handicapping.

Once I get to the point where my probabilities are ACCURATE, then there is absolutely no point in working any more to improve my handicapping. If the probs are accurate, I'm done with improving my overall handicapping skills.

The natural question is, "how do you know the probabilities are accurate?" Well, you cannot test any one race for accuracy. What you do is to handicap a lot of races, and test your overall statistical accuracy. And you do it in a way that gives you great security that the method you are using IS accurate... for EVERYTHING you are doing.

Here's how you do it. First, you use some method to handicap a fairly large number of races. Obviously, the larger the sample, the better you know the accuracy of your handicapping. For now, let's say we handicap 1,000 races. Then you pick out any ONE probability. Let's say we choose the probability that the #1 runner will finish in first place. You take your probabilities for that cell in all 1,000 races, and add up those numbers. (Like .1341 (from above) + .2341 (from the second race) + .1254 (from the third race) and so on. Just add up all 1,000 probabilities that you have calculated. You'll end up with some number, like 156.7342. What that tells you is, for those 1,000 races, according to your probabilities, the #1 runner should have won about 157 times. Then you simply add up the ACTUAL number of times the #1 runner won in those 1,000 races, and compare it to your sum of probabilities. The closer you are, the more accurate your overall handicapping.If you find this hard to grasp, think of it like this. Let's say you have a runner in a race with a 25% chance of winning. That's 0.25, right? Now let's say you have four such races, with one runner in each with a 25% chance of winning. When you add those up, it's .25+.25+.25+.25=1.00. So you would have expected, in four races, that your #1 runner would have won once, on average. (Just what you would expect from 3-1, right?) If you DID have one winner in those four races, you know you're on target, and your probability estimates are right! Cut me some slack on the small sample size. This doesn't really work well with 4 races, but it DOES work just fine with 400 or 4,000. My method is extended to measure dozens of probabilities across hundreds or thousands of races.

Now, that tells you how accurately you have built probabilities for the #1 runner winning. But you should do the same thing for EVERY ONE of the probabilities you have calculated. So I add up the probabilities I calculated for the #5 runner finishing in 8th place, and then check that number against the REAL number of times the #5 finished in 8th place. And so on. You end up with a big matrix that shows you exactly how accurate your handicapping really is.

To measure your overall accuracy, you use a statistical technique called "variance." (Check out any beginning statistics book to understand variance.) Basically, variance measures how far your probabilities are AS A GROUP IN TOTAL from the real results. (Basically, it's the square root of the sum of the squares of the differences between your expected probabilities and the observed percentages.) So you can calculate the variance of the probabilities from the actual results, and you will get some number for that calculation.

Then the exercise becomes to continually try to improve your handicapping. As you improve, the variance will go steadily down. In a perfect world you would keep working at improving your handicapping until your variance is zero, but that's not possible, of course, in the real world. So you work at it until you get to a variance that's reasonable, or until you can't find any more ways to improve your handicapping.

Now another cool part of this is that you can use the exact same technique to measure how accurate the CROWD'S handicapping is, and compare that to your own handicapping as a benchmark. Just change the post-time odds into probabilities, and use the same method! Of course you can also test different handicapping TECHNIQUES to see what works and what doesn't.
Now all this is based on the idea that you can only get so good at handicapping. After all, races are governed by probabilities, and not certainties. Every single runner that races has some real, finite chance of finishing in first place. And a chance of finishing in second, and in third, and in all places including eighth. Granted, they may have a BETTER chance of finishing in one spot or another, but they have SOME probability of finishing in any position.

Further, you must understand that no amount of handicapping can ever get you to the point where you can know perfectly how any given race will finish. You can make projections based on your handicapping, but you cannot KNOW how each race will finish. Your goal becomes to minimize your variance across a very wide range of races, and individual statistics. Once you get it to a minimum, there's no reason to continue trying to improve your handicapping.

So, how do you know when to stop? Actually, that's very easy. Probabilities are a "zero-sum" game. That is, if you look at all possible outcomes of something, their probabilities have to add up to 1.000 or 100%. Let's say you handicap a race, and assign win probabilities to all the runners. If they don't add up to 1.000, you've done something wrong!

But let's say the DO add up to 100%. Your handicapping can now be either good or poor. Let's say it's poor. You've given a runner a win probability of .33 when it SHOULD be .25. You say he'll win more often than he actually would for real. Well, because probs are a zero-sum game, if you OVERESTIMATE one runners chances, you must have UNDERESTIMATED one or more OTHER RUNNERS' chances!

Let's relate this back to the variance method above. Let's say you handicap 1000 races, and your probabilities say you should have the highest-probability runner winning 281 races. If they ACTUALLY won 241, you've overestimated their probabilities. Well, if you look at the OTHER finish positions, you'll find that one or more of them must have been UNDERESTIMATED. So to determine if your handicapping is as good as is possible, you keep working until the differences ON ALL POSSIBLE FINISHES are roughly equal. If you are off by about an equal amount on every possible probability, then you know that your handicapping CANNOT GET ANY BETTER! If, however, you have one or a few segments that show higher and lower variance than average, then you know your handicapping can still improve some.

Amen

[Duritz]

Thanks for that.

Have you done analysis to see what the crowd handicapping is like? As you said earlier, the crowd gets it right odds wise but make stupid mistakes, have you tested their win odds and seen if they are basically accurate with that?

Duritz.

[crash]

The best crowd average is 30% winners for a LOT of -15%.

[Duritz]

Well the crowd have to lose 15% b/c that's the TAB take. What I mean is, are their selections winning the amount they should be.

[crash]

I meant the 'crowd' is loosing 15% on their bet outlay based on their overall best average win SR [30%] and the prices returned for the win after TAB take, not before it.

[foxwood]

Hi Woof,
You said:
Quote"3-dog box: 80 hits out of 1286 races (6.22%), average payoff is $89.66, and the ROI is a miserable -53.52%. (BTW, the 6.22% hits is very reasonable for a three-dog box.)

The HANDICAP isn't important. The WAGERING means the world! Just LOOK at that spread. The BAD wager has a 53% LOSS and the GOOD wager has almost a 63% PROFIT! All based on the exact same handicap.

In all my years at the track, I've never seen ANYONE that can improve their HANDICAPPING from a 53% loss to a 63% profit. But you CAN improve that much if you study wagering." endquote[/QUOTE] (I still haven't mastered the art of quoting)


Try as I might I can only make that a ROI of -7.04%. If the average payout were exactly half of your $89.66, i.e $44.83 then the ROI would be -53.52%.

I don't wish to sound patronising or sycophantic but your contributions to this thread have been the most enlightening I've seen. Many thanks

[woof43]

Many thanks for the kind words.

[lomaca]

[QUOTE=woof43]Duritz,
This is where Handicapping, pricing (assigning probabilities) betting overs and winning starts.

Woof43 Hi!
I read your reply to Duritz with great interest.
Good stuff!!!
I have studied statistical analysis, (have to admit not willingly it was just a subject we had to do), and never thought of putting it to good use, until I read your post.
Always tried to improve my ratings and now I realise it is nearly as good as it gets (or as good as I can make it) and now I have to find a way to make it more profitable. Sometimes we need just a nudge in the right direction to go forward.
Thank's for your input.
Ps. Duritz! If you read this, my apologies for the use of parentheses TM(Duritz)

[Sportz]

lomaca,

Could you please read my reply to you on the February Tipping Comp thread.