Thirsk 4.55 17th May 2008

[Matunty]

MY 'SYSTEM' says Ishetoo in the 4.55 at Thirsk. I would like someone to communicate with me about how I have selected this and then criticise my 'system' in order that I understand whether I am on the right tracks. My reasons are these: Please don't laugh if you are a pro. Your response to this will help!

I have selected Ishetoo in the 4.55 at Thirsk because:

1. It is listed 5th and I have awarded 3 points
2. It has -2 points because it is 2 away from an RPR comparison table that I use based on other advice received here This separate excel sheet awards points for slected courses of which Thirsk is one
3. It has an RPR of 115 (so I add 15)
4. 5 points added because t is both Course and Distance (and the only one in this race). I give 2 points for C and 2 points for D but together 5 if the runner is the only CD in the race
5. The ground is Good to Firm (1 point) I only back horses where the grand is either Standrd, Good or Firm (and combinations of these G-F etc)
6. Add 2 points because the horse is within 2 years of the age of the race i.e. Ishetoo is a 4yo in a race for 4 year olds
7. Add 1 because it was 3rd last time out
8. Add 1 becuae it is the 3rd top weight in a handicap
9. Add 3 using a handicap table for selections against fav in the betting which I found elsewhere on win to win and have adopted this table as another variable.

If you are laughing I would like to know why as this would really help. It is 16.19 and I have deliberately written this before the race happens!

Using my point award system against these factors, the maximum score is 47, the lowest possible is 2. My rule is to back horses where the 'system' awards 25 points of more. I never back if 24 or less. Ishetoo has a score of 30 so I have backed it.

Should I take up darts? :yikes:

Steve

[mathare]

OK, here goes nothing...

I have selected Ishetoo in the 4.55 at Thirsk because:

1. It is listed 5th and I have awarded 3 pointsListed 5th where? 5th in terms of what? It could be distance the head boy lives from the nearest brothel for all the info you've given

2. It has -2 points because it is 2 away from an RPR comparison table that I use based on other advice received here This separate excel sheet awards points for slected courses of which Thirsk is one It's two away from a comparison table of RPR against what? Why only selected courses?

3. It has an RPR of 115 (so I add 15)To the RPR? To the number of points allocated by your system? Why 15?

4. 5 points added because t is both Course and Distance (and the only one in this race). I give 2 points for C and 2 points for D but together 5 if the runner is the only CD in the raceHow did you come up with the number of points to award for C and D wins?

5. The ground is Good to Firm (1 point) I only back horses where the grand is either Standrd, Good or Firm (and combinations of these G-F etc)Is the ground really only one-fifth as important as the fact the horse is a CD winner? What evidence do you have to support this? Why don't you award points for other goings rather than just forget the bet altogether? It doesn't make sense to have a points system then just ignore certain goings does it? Award fewer points or take points away but don't just scrap the race because of the going after awarding points for all the other factors

6. Add 2 points because the horse is within 2 years of the age of the race i.e. Ishetoo is a 4yo in a race for 4 year oldsWhat? So what happens in an any ages race? How can a horse in a 4yo only race not get these points? And why do you think age is twice as important as going?

7. Add 1 because it was 3rd last time outThat seems fairly random. Why 1 point? What does it get for finishing 7th, for example?

8. Add 1 becuae it is the 3rd top weight in a handicapSee above. What if there are only 3 horses in the race?

9. Add 3 using a handicap table for selections against fav in the betting which I found elsewhere on win to win and have adopted this table as another variable.Is this other variable suitable for this system? Why have you adopted it?

If you are laughing I would like to know why as this would really help.I'm not :laugh, more :headbange and :splapme and :doh

Using my point award system against these factors, the maximum score is 47, the lowest possible is 2. My rule is to back horses where the 'system' awards 25 points of more. I never back if 24 or less. Ishetoo has a score of 30 so I have backed it. How is the maximum 47? How is the minimum 2? Where did a cut off of 25 come from?

I really don't understand this system at all. Why have you allocated points in the way that you have? Have you done any research into which factors influence racing results most? If not, how can you weight the various factors as you have? You give 1 point for a right ground but 15 for the RPR - is the RPR 15 times more important than the going? You have several factors in there multiple times. RPR is used in factors 2 and 3. Things like performance LTO, CD wins etc will be built into the RPR calculations. You use weight in handicaps and a separate handicap v odds table effectively using the handicap twice in different ways. There's no real rational reasons for this system to work as far as I can see. What back-testing have you done on it?

I just don't get it at all.

[Matunty]

Oh well. Not bad. Ishetoo was going well, and seemed to be holding a little back for the final furlong until Manzilla came up to join and then romp home. So Ishetoo was a good second. Money spoke at Thirsk on Manzilla who went from 12/1 to 13/2 though I am not sure how much money is needed (relative to the course and occasion) to reduce the odds to this extent?

My main question is, does the respectable second confirm the 'system' I introduced at the start of teh thread? Or did Ishetoo come second for a whole load of other reasons?

Any criticism (constructive( would be welcomed.

Steve

[mathare]

My main question is, does the respectable second confirm the 'system' I introduced at the start of teh thread?One swallow doesn't make a summer and one loser certainly doesn't validate a system. Nor would one winner. You need a LOT more data than that.

[Matunty]

Thanks for your reponse to this. I respect your criticism and will reflect upon it. I don't want 'others to do all the work for me' which was what you suggested in my Bath discussion. I am happy to help others but as you can se I am a novice who wants help. Is that so bad? Thanks for your comments. I think your main criticism is not that the factors I ahve chosen are wrong, but that the weighting is wrong. I accept that and will work on this. So, taking the risk that you won't be offended, how would you rank(weight) the factors that I am using. If you think this is too much of a secret then find.I will experiment myself.

All the best and again, thanks for your help and time.

Steve

[mathare]

I think your main criticism is not that the factors I ahve chosen are wrong, but that the weighting is wrong. I accept that and will work on this. So, taking the risk that you won't be offended, how would you rank(weight) the factors that I am using. The factors may be wrong but there is no easy way of telling. To get this system working would be a huge job, in my opinion.

What I would do is something like:
Set up a spreadsheet (or similar) allowing me to record points for each factor separately, along with race result, SP etc
Apply the rules as written above to as many horses as possible
Add the results and SPs to see what sort of SR and profit is generated by the current rules
Tweak the weightings, drop rules, add rules etc as necessary and in each case see how that changes the SR and profit
Like I say, not a quick job. At the same time I would be reading all I could about racing and trying to develop my own understanding on how important trainer, jockey, weight, distance, course, going etc seem to be in determining who will win any given race.

I don't have any desire to develop my own systems but if I was starting from scratch I'd do as I just stated. Actually, that's a lie. I'd get RSB or RaceXpert or something and put in serious amounts of time learning that software and understanding how to get the best out of it.

[Street cry]

I am really struggling with this seemingly senseless rationale.
No offence but in short you are wasting your time

[Street cry]

Start with a list of sensible variables and comments on why they are included

[Matunty]

Thanks for your response again Mathare. I am interested in developing a system. I realise I have a lot to learn and a lot to lose! I will examine the weightings first as your 'arguement' hear is strong. It was stupid to award 15 to

[Matunty]

stupid to ... award RPR with 15 and ground 1. I realise that. There are some basic questions tho and I will endeavour to read more or post these basic questions in order to make my system more in the right direction; How should I rank these factors (in order to weight them). Or, if a pro was to put them in order of their influnce on the race I wander what he/she would do? time elapsed since last run, course, jockey, weight, position last time out, Course winner, distance winner, ground, weather conditions? !, allowance (of jockey), RPR, OR, TS, draw, age, type, (seller, maiden, handicap, group race etc) etc etc etc. THERE MUST BE POTENTIALLY hundreds of other factors not mentioned so where does the amateur (who is keen like me) begin.?

[Street cry]

You begin with shall we say common sense Variables.
Lets start with favs in non handicap races over jumps that gives around a 40% strike rate
and - 3.5% poi

If you then move to look at just geldings then you increase sr by 1% and increase poi to -2.5%

Now we ensure that we are only backin these favs when they are down in class that improves things to 45% strike rate and 2.8% poi

if you use chase races only and avoid all hunters or maidens you improve things further to 50% strike rate and 4.66% poi.

The above is profitable over the last 10 years on betfair but is too marginal to use.
the point is that you want to be looking at variables that actually mean something not complicated non effecting approaches such as comparing rpr etc etc cheers sc

[Matunty]

I respect the patience given to be by those more experienced. Thanks

I am still learninga nd am grateful for all the advice I can get.

Steve

[Street cry]

as i say make a list of variables you think are worthy of inclusion and why and post em on here

[Matunty]

I am learning all the time. I jave deleted some variables (like RPR) which has largely been discredited. I have altered the weightings and started to look at the % success in the type of race in question of both the jockey and the trainer, making sure I look at the % for the whole of the previous season (2007). I have awarded a greater weighting to Good over firm, standard and soft etc (in fact I am now avoiding any race where the ground is anything other than standard, good, good to firm. The Boss suggests that some courses list firm when it is actually hard! Anyway. I am reading system builder and this is informing the development of my system. Win or lose, its more creative than the crap on tv these days! Unless you have Racing UK!