Assuming you didn't find a tree and a rope - at least one of your seven losers was not a valid selection according to your rules. (Can't remember which one but I think it was in a maiden with lots of horses that had never run.)
Your system is logical and works - just needs some filters. Eg drop the class 1 & 2 races which never pan out.
I'd have a careful trawl back through your results and see where the winners and losers are coming from: big fields? tiny fields? soft ground?
BORNLOSER..........I did say many posts ago that I WOULD include some MAIDENS/etc. in SPECIAL CIRCUMSTANCES........The rule that it breaks is HAVING TO HAVE FIGURES FOR EACH RUNNER.....In some Maidens/etc. you may find one horse not having run/etc so will not have a rating of any sort, but maybe forecast price 66/1 or 100/1...and the system selection is say 6/4 forecast favourite.....I take a chance that this one runner will not win and still use the race.So far none of these rags have won.
FROM TODAY I WILL STICK TO THE RULES EXACTLY AS PREVIOUSLY POSTED TO SEE IF IT HELPS.......i.e....EVERY RUNNER MUST BE FULLY RATED.....TODAYS SELECTIONS.....3.50 C.....STANLEY WOLFE......4.20 C........HYPNOSIS.......3.30 L.....BAIZICALLY........6.45 T......MOLLYANKO.
FAO VCAMRN..............If its possible could you put the 4 basic rules of the system through your magic box.TOPRATED MASTER RATING(OR JOINT).........LTO SAME RATING........ALL RUNNERS MUST HAVE A RATING........LTO MAX OF 28 DAYS.............I suspect rule #3 may be a problem, if its a hassle forget it...CHEERS.
Because of data validation issues as posted in the HSL3 thread it will be a few days before I'm ready to crunch some data from Jan onwards. I can do the rule that all runners must have a rating but I have to manually tweak it. BTW as you well know from 'THE OTEHR PLACE' only 5% on newcommers win any Maiden as well as the same number again win that did not palce LTO. This is the reason I don't really care about ignoring all those with no speed ratings.
VC Thanks for the reply and help.I just thought I had better find the BASE FIGURES for the SYSTEM RULES EXACTLY AS I POSTED THEM.......shame there is no magic button here to do it in 3 milliseconds here.
YESTERDAYS RESULTS..........L 14/1, 2nd EVENS, L 5/2, L 4/1, ........R/T.........25/79.........+ 3.2 POINTS............I hope you can see now why I wait for 200 selections before I reach for the wallet.!!................TODAYS SELECTIONS.........3.30 L.......JOY + PAIN..........7.30 W......DEVINE DANCER.
You finally got a winner. That was a good horse and I myself was on it for different reasons.
According to the modified rules I posted which does not care about unraced in the field, it found a nice winner in the 7:00 Windsor today which was La Roca at 5/1. That makes it 3/7 this month with +16 points.
I hope I can crunch out some numbers using your rules tomorrow to see what it looks like from Jan06 to date.
VC.........I will keep on posting the BASIC SYSTEM SELECTIONS till you have time to do a BACKTEST if possible............YESTERDAYS RESULTS..........WON 15/8 + L 33/1.........R/T........26/81........+ 4.08 POINTS...........TODAYS SELECTIONS.........1.50 L.....AGGBAG.......5.10 N........ZAMALA.
YESTERDAYS RESULTS.......L 3/1, 3rd 7/2, 3rd 6/4, L 7/1..........R/T........26/86.....30%......LOSS of 0.92 POINTS...........And as we have gone into the red I think thats the end of the trial for me.Its obviously a PLUS FACTOR using this method ( as you are beating random selections by about 25%).But its never going to get + 30 % ROI which I look at as a minimum .
Good work Dylan and nice to keep the trial going for a longish period to test the highs and lows.
Some good selection criteria has been added to all our knowledge base which is invaluable.
Small point though I think 30% ROI is asking a bit much, I look to have a bank of systems/methods that long term offer about 10% ROI.
I am not a big gambler and with these selections on the average day turnover about £250 quid that is roughly £90k a year so looking for income of £8-10k a year from racing.
Still difficult to acheive but I think a spread of systems rather than looking for the holy grail it is more acheivable.