has anybody tried developing a system based on "orphan winners" (those horses that are the only last start winners in the race)...

i read an old article in ppm a little while ago by a lady called janice crawford, and she said an analysis of orphan winners at tab meetings for a year was conducted....it showed by just backing orphan winners at level stakes a pot of about 2% was achieved over more than 1,000 races....

it set me thinking, that with a few filters a profitable system may be achieved...

i tried - horses with the letter "c" or "td" after their name and pre-post favourites, but neither lifted the pot much...ppf's are currently showing about 7% pot after 150 races, the others haul in some nice price winners but not enough to lift the pot above about 5%...does anybody else have other ideas???

If I understand you correctly that you are considering betting on those horses which won their last outs, be aware of whether they have moved up in class and picked up additional weight because of it (as is done in the U.S.).

My horse won two weeks ago in NW2. He ran yesterday, handicapped by nine additional pounds (and another 1/16 of race) against more proven horses carrying 5-8 pounds less (because they had not won lately).

If the conditions affecting the horse did not change, your idea might work. However, there are many other reasons a horse may "bounce" (not run back to the previous start).

It would be very interesting to know what the strike rate was, the longest run of outs and the average dividend paid. I level bet, but someone who is into staking could probably make something out of it, depending mainly on just how short priced these horses are and how big the gap between wins.

Hello again Noel, I spent an hour going through Wed/Sat/Sun results for approximately 8 weeks and found that yes, it did show a small profit. There were no really big priced winners in the sample (my idea of a big winner is over $9) and there were certainly days when you lost on virtually the lot. The only way to make it pay in its current form is to bet on every race meet both metropolitan and provincial, and in fact the horses with the best payout often were in the provincials. I have a couple of thoughts on improving the strike rate. Firstly eliminate those races where not all the horses have exposed form. Quite a few of the selections were for 2 and 3 year old horse races where some of the horses were having their first race. Secondly it might be a good idea to eliminate those horses who are inconsistent. By this I mean there were many selections where the horse had indeed won its last start but in general had a woeful track record. There are always fluke wins and if you want to cut down on looking at form, class and weight, then eliminating these horses probably does some of that for you. There were plenty of races where I would have discarded the last start winner based on its last five starts, however I didn't have the time to work out whether this would have made a difference or not. If you get a chance to do the figures, or have any ideas let me know. My email address is : luckylouie@bigpond.com.au
I'd be interested in seeing if we could make it a decent proposition.