• Hi Guest Just in case you were not aware I wanted to highlight that you can now get a free 7 day trial of Horseracebase here.
    We have a lot of members who are existing users of Horseracebase so help is always available if needed, as well as dedicated section of the fourm here.
    Best Wishes
    AR

The Ides Of March

Chigorin

Chigorin

Filly
Hello everyone!

I've had my head down all winter, further developing my systems and methods and have had some half decent results. To bring to an end my long absence from these forums I thought I launch a new thread and since today is the 1st of March I've named the thread after the day in the Roman calendar associated with the overthrow of Ceasar and the end of the Roman Republic. (Strictly speaking I think the Ides actually refers to a day in the middle of the Roman month but poetic licence and all that.)

So, the idea is to make a post every day this month if I can. Each post will include a tip from my new system and a brief comment about some point of learning that has come my way in its development. Some of these points may be obvious to you but hopefully there will be something to help those who are interested in system development.

Anyway, more of that tomorrow. Here are today's tips, rushed out as the first race is about to off.

Lingfield 14.40
Luis Vaz De Torres 27
Searchlight 22

Lingfield 15.40
Jazri 36
I'm Harry 17.5

Lingfield 17.10
Candesta 12.5

(Betfair prices I have taken today)
 
Good morning and thank you for the "Likes" gentlemen.

March 2nd

The concept for this thread was to recap on some of the thoughts, observations and conclusions I have gathered from a monastic winter of system study in the hope that they will be of interest, or better still generate some dialogue. Although I haven’t planned ahead, I envisage that by the end of March a (somewhat) coherent batch of ideas will have been presented that will outline a road to system development.

My starting point then is a study the notion of “Is It Possible?”

If you are going to develop a profitable system then, before you even open Excel, HRB or take the lid of your biro, you must believe that not only your task is possible, but that you are going to do it. If you haven’t created a profitable system up to now, then even before you ask yourself “why?” I suggest taking a few minutes to mentally frame what you are setting out to do.

I know that it is possible to develop a winning system.

This is probably not what many of those around us (especially outside out betting fraternity) will try and tell us when they express their belief that the bookie always wins and holds all the cards. Before we begin to construct a system let’s remind ourselves of just one of the cards that we, the punter, hold (although I think there are 6 distinct advantages that we have, which I call the six silver bullets).

Our primary advantage is that we choose if and when to bet. This is huge. The bookie is obliged to put (normally a selection of) odds up on every horse for every race, all day, right up to the start (and beyond if you use the exchanges). Like a lion waiting for a sign of weakness or lapse of alertness in the wildebeest herd, we just have to wait for our moment. When we see one mistake, one point of critical weakness, it’s then that we make our move.

Put more mathematically, the combined actual winning chances for each horse in any race must equal 100%. With skill, if you time your bets well on the fluctuating exchanges, you could actually cover each horse so that the combined winning chances equate to less than 100%. Leaving this possibility aside and dealing with BOG prices which might typically be around 105% in total, we can see the scale of our challenge. To guard his advantage the bookmaker normally gives himself a “mark-up” on what he judges to be each horse’s winning chance. On average this mark-up must be of less than 1% of what he thinks is its “real” price. Where we are confident that he is wrong, even allowing for his protective mark-up, we should bet. Of the dozens or even hundreds of horses running each day we can give him the benefit of the doubt as many times as we wish. But be sure that in many, many of these cases the bookmaker is a long way out with his estimations. Furthermore, often he knows this. The price the bookmaker trades at is as much determined by demand for that price as his conviction that it is accurate. If the market backs a particular runner the bookmaker will reduce the price, but not because he believes he has misjudged it (and correspondingly the prices of some other competitor will ease a fraction irrespective of what the bookmaker believes are "real" chances). Thus it follows that much of the time we are not even pitching our wits against a team of professional experts who may be privy to inside information, but we are actually trying to out-do other punters. This is vitally important. If we are going to fare better than the masses then we must have some advantage over them. Something which sets us apart from them.

Actually I believe we need two things. We must have a system that identifies bookmakers gross errors more accurately than it does not, and secondly we must have (what I have called elsewhere) a process for using this system to bet.

If you don’t believe this is possible then please identify yourself as a “recreational gambler” and enjoy your hobby. However, if you do think it is possible to beat the bookie please keep reading.

Bets I have made this morning (all Wolverhampton):

14.55 Teepee Time 110
15.30 Halstaff 18
16.05 Miss Buckeroo 34
16.40 Auspicion 30

Yesterday 5 bets, no winners. (Two seconds including a photo)
Total to date Bets 5, Profit -5
 
Good afternoon fellow punters. Thanks for taking an interest and especially those of you who have taken the trouble to like or comment. Now to continue our investigation into system building....

March 3rd

Yesterday I tried to impress the idea that a profitable system is a real possibility and introduced the conceptual cornerstone of system building; namely that in some instance we must more accurately assess a horse’s winning chances that the combined efforts of bookmaker and market, and we must “know” when we have done so.

With apologies to those readers who are eager to “get on with it” and start creating our system, I wish to spend another day pondering a fundamental preliminary question. Assuming you are not operating a long term winning system, that is to say, making clear profits from your bets, then I suggest asking yourself why haven’t you succeed in this task before now?

I expect that the answer is one or both of two things:

The most obvious possibly is that the beliefs, assumptions, methods of calculations etc that underpinned your system were false or inaccurate. That is: your system doesn’t identify enough well priced winners.

The second possible failing was that you simply failed to put your system into operation properly. That is to say that you may have been accurately identifying errors in the bookmaker’s prices, but “bad luck,” rushes of blood, over-cautiousness, lack of scrutiny or some other such “human error” has resulted in a loss, or at least lack of steady profit. This second type of failing is what I call a failure of process rather than of system and is a very real enemy in an activity so emotive and psychological as gambling. Far too little attention is paid to process and at the risk of furthering this error and we will return to this subject later in the month. I shall, for now at least, focus on system.

Let us suppose your failure to make profits from horse race betting is because your system is not good enough to beat the bookies. How can this be? I ask you. Were you forced to make any of the bets you have made? Did you not test your system on an almost infinite pool of races?

Clearly your answer to both these questions must be “no”. Perhaps you have made the mistake of betting using a system that has not undergone rigorous enough trialling (this is a failing of process). Perhaps you have been better than this and not bet at all because, as yet, you have not found a system to be hypothetically profitable.

In either case you need to work harder on your system. It is only your sticking at it that is between you and success. It’s like an incredibly complex Sudoku puzzle. No matter how difficult, the answer is there to be uncovered. The clues are all there. You are not looking at a puzzle that cannot be completed.

Your failure to make money from betting is no more than lack of application on your part. If you are prepared to put in the hours, with the right mind-set I genuinely believe you can unravel these clues and make money.

Tomorrow we shall examine another fundamental concept that we must understand in order to construct a winning system.

Today's Tips (all Southwell and Betfair prices I've taken this morning)
13.45 Feelin Dicky 30
14.50 Galvanize 30
16.00 Prayer Time 36

Yesterday's Results
4 bets, 1 winner. (@34) Daily Profit (after commission) 28.35
Total to date Bets 9 , Profit 23.35
 
Welcome back from your hibernation Chig i thought the Jungle was hot.;) Any member who thinks about what you have written should find it most beneficial.
 
Hello all, and thanks for your interest in this daily series of reflections on lessons I have learned while developing what I hope continues to be a profitable betting system.

March 4th

Thus far I’ve argued that to construct a profitable system is possible – but is hard work. In other words I've listed two facets, namely belief and effort, which may best be described as qualities of right attitude. Today we move to the sphere of knowledge Most of you won’t have found what has been claimed so far contentious but some of you may find today’s notion more opaque.

My claim is that certainty does not exist.

This is true not only at the sub-atomic level (i.e. Heisenberg’s Uncertainty Principle) or at a global level (the butterfly effect or chaos theory) but in the world of horse racing.

When the stalls are full and the flag is raised, no one horse is fated to win. Nor is the outcome random. A probability field exists such that were the race run 100,000 times one horse would win more times than any other, and one horse less than all others. This probability field, or wave, breaks down to an actual event – and outcome - only when the race actually happens.

The market, as presented by the bookmaker, is an imperfect probability wave defined by the bookmaker. Imperfect because, as we have discussed, it does not represent EXACTLY a 100% likelihood. Because we are interested in spotting failings in the market we need to be able to estimate the winning chances of each horse. It will not be possible for us to identify one horse as having more chance that the market suggests until we have completed our own probability wave and assigned a winning likelihood to EACH horse. This is because until we have calculated a chance for each of the runners we cannot know how the other horses are likely to be effected by the 'missing' horse.

Our job is to estimate a complete probability wave more accurately than the market and we shall take this a step further tomorrow.

Today's Tips (all Wolverhampton and Betfair prices I've taken today)
18.15 Comparinka 36
18.45 Silk Bow 36
19.15 Don't Tell Louise 100

Yesterday's Results
3 bets, 0 winners. Daily Profit (after commission) -3
Total to date Bets 12 , Profit 20.35
 
Thanks again for your comments and acknowledgement. Today we continue with our discussion of probability waves. If you are new to the thread you may benefit from reading some of the previous entries.

March 5th

I believe that the concept of probability waves is central to system construction. However, the difficulty with any examination of the probability wave relating to a horse race is that we only get to test our hypothesis once. This can tell us next to nothing with certainty. Horse Alfie winning the actual race does not tell us how many times Alfie would have won had the race been run 100,000 times. It may have been that Bertie would have won the other 99,999 times.

Yet we need to test our hypothesis if we are to have the confidence to bet on the basis of it. Thus we must run a system hundreds, preferably thousands of times. This way the outline of the system’s effectiveness becomes manifest even if we can never do more than infer its accuracy when applied to any one race.

If we win 100 races in a row, each with a horse we estimated to have a 25% chance of winning, it follows that our system is not good. We may have won 100 races in a row and now be filthy rich but the truth is that we have grossly underestimated our selections’ chances.

It is only from a large field of results that we can draw meaning from analysis of probability waves. And we should note that even here, with a large pool of data at our disposal, we are not any more certain about the outcome of any particular probability wave. Our growing certainly is in whatever system we use to estimate that wave.

Tomorrow we will talk some more about the process involved in developing a system to estimate probability waves.

Today's Tips (all Betfair)

Lingfield 16.15:
Tutu Nguru 22
Above And Beyond 38
Squash 28

Yesterday's Results
3 bets, 0 winners. Daily Profit (after commission) -3
Total to date Bets 15 , Profit 17.35
 
No tips today as there is no flat racing - which I think make today the first day I don't get a runner up! Meanwhile we will continue with my considerations relating to constructing a probability wave which is the center-pin of our system.

March 6th

To draw our probability wave for any race I suggest a two stage process – although each stage will be subdivided into numerous parts.

As probability waves are numeric we need to establish a score, or score range for each horse running and then we need a system of translating these scores into each horse’s percentage chance of winning such that the total probability of all the runners equals 100%.

Today we will begin our look at the later stage – using a score to ascribe a % probability - which should be simpler although perhaps not simple!

Let’s imagine we are studying a 3 horse race involving Alfie, Bertie and Charlie. For now let us assume we know nothing about any of the horses (or the course). We have no reason to favour any one horse over another and would therefore construct a probability wave where each horse is ascribed a 33.33% chance of winning.

Dissatisfied with such a lack of knowledge we look up the horses Official Ratings which are as follows:

Alfie 45

Bertie 50

Charlie 55

We must now decide a basis by which we will translate the difference in each horse’s score into chance of winning. We might decide, arbitrarily, that each additional point in rating equates to an additional 5% chance in winning. Thus Bertie has 25% more chance of winning than Alfie (5 x 5%) and likewise Charlie has 25% more winning chances than Bertie. We can quickly see that on this basis of allocation our probability wave shall be:

Alfie 8.33%

Bertie 33.33%

Charlie 58.33%

Tomorrow we can see if we can improve on this beginning. It would be great if anyone following this thread had any suggestions.

Yesterday's Results
3 bets, 0 winners. Daily Profit (after commission) -3
Total to date: Bets 18 , Profit +14.35
 
Chigorin said:
It would be great if anyone following this thread had any suggestions.
Click to expand...
Enjoying the thread but for me its a little too early to start making suggestions as I do not have a clear understanding of where you are taking us and do not want to potentially derail your thoughts as you seem to have something mapped out that you wish to deliver. Apologies if I have the wrong end of the stick!
 
Chig i think ArkRoyal ArkRoyal makes a valid point above.How about giving us an agenda or time line so that we can confine questions or suggestions to your current sharing and not unintentionally cause disruption.

Re your today's example you do not state if the three runners are contesting a hcap.If so in theory weight will equalise your improved probability %. ?
 
again not sure what the question is - there are a few ways of creating Odds line I posted one yesterday and here another I found earlier today whilst supposedly preparing for Mothers day

In the software section IS author trecelyn second page of the thread
trecelyn said:
I have uploaded an excel spreadsheet that calculates odds based on inputted ratings. Not sure if it's working ok, if not pm me and I will email it to whoever wants it.
Click to expand...
has an odds to probabilty /% odds calculator
 
Last edited:
Chigorin said:
No tips today as there is no flat racing - which I think make today the first day I don't get a runner up! Meanwhile we will continue with my considerations relating to constructing a probability wave which is the center-pin of our system.

March 6th

To draw our probability wave for any race I suggest a two stage process – although each stage will be subdivided into numerous parts.

As probability waves are numeric we need to establish a score, or score range for each horse running and then we need a system of translating these scores into each horse’s percentage chance of winning such that the total probability of all the runners equals 100%.

Today we will begin our look at the later stage – using a score to ascribe a % probability - which should be simpler although perhaps not simple!

Let’s imagine we are studying a 3 horse race involving Alfie, Bertie and Charlie. For now let us assume we know nothing about any of the horses (or the course). We have no reason to favour any one horse over another and would therefore construct a probability wave where each horse is ascribed a 33.33% chance of winning.

Dissatisfied with such a lack of knowledge we look up the horses Official Ratings which are as follows:

Alfie 45

Bertie 50

Charlie 55

We must now decide a basis by which we will translate the difference in each horse’s score into chance of winning. We might decide, arbitrarily, that each additional point in rating equates to an additional 5% chance in winning. Thus Bertie has 25% more chance of winning than Alfie (5 x 5%) and likewise Charlie has 25% more winning chances than Bertie. We can quickly see that on this basis of allocation our probability wave shall be:

Alfie 8.33%

Bertie 33.33%

Charlie 58.33%

Tomorrow we can see if we can improve on this beginning. It would be great if anyone following this thread had any suggestions.

Yesterday's Results
3 bets, 0 winners. Daily Profit (after commission) -3
Total to date: Bets 18 , Profit +14.35
Click to expand...

Try Page 16 in the Smartsig Book by Steve Tilley of Chi Fame
 

Attachments

  • 0606-mm7g-4.pdf
    1.4 MB · Views: 42
Many thanks for all these inputs.

Apologies for the rather vague way I invited "suggestions" when I have not set out clearly enough my bigger picture or what more precisely I was inviting suggestions about!

mick mick , sorry, I've failed to make it clear that weights don't come into the very hypothetical question I am setting here (you are quite right that they would in real life). I am asking how, once we have given a horse a score, does this score relate to their % chance of winning (irrespective of what went into making the score). My first suggestion today was that each point of score equates to a 5% greater chance of winning.

I think Chesham Chesham has understood this, for since he has offered two fine alternatives to the question of how can we translate a numerical advantage in score or rating into a % chance to win. I've just had a quick look Chesham Chesham and these are of great interest - thanks!

Hopefully what I am attempting here will become a little clearer tomorrow and as I go on. I realise I'm being pretty ambitious in trying to set out relatively complex questions with little explanation and in a very short space. The feedback you have ALL given is VERY helpful because without it I have no idea of whether what I am saying is making any sense or possibly too obvious to be of interest. I wont say more now because I have tomorrow's installment planned but please know I would very much like to be told if what I am saying is not making sense.

Thanks again
 
Hi Chigs cheers i understand your point of score concept.The difficulty i find with this is establishing the correct weighting for each aspect used in the scoring.I look forward to reading your thoughts on this.
 
You must log in or register to reply here.
Back
Top