What Happened To Systems?

I think I’ve told this story before, but maybe worth repeating.

Many years ago I had a season long nap of the day competition with a work colleague - loser pays for a slap-up lunch. The boss caught us studying form one day and asked if he could join in. He knew (he said) very little about horse racing so he would just have the third horse down, in the third race of the day at the main meeting. We both finished on the plus side, but had to buy him lunch two years running!
 
I used to be interested in systems.

The logic in creating the system is what interests me, not really following systems themselves, unless they become immensely profitable!

Sometimes one good hunch based on a feeling at any given time of a season can turn in a decent profit.
 
Everyone can have their own view on everything but personally I prefer the term selection process cos I don’t really distinguish between method and system other than a ‘system’ tends to mean it’s more open to automation and more fixed, however who is doing the automating makes a big difference as to what is possible i.e. a professional vs a have go hero who’s recorded a macro in excel.

But I'm somewhat sceptical of any method/system which wouldn't necessarily come up with the same selection if you looked at the race again with exactly the same information available on paper, on screen or in your head etc whether before or after the race has run.

I see a selection process as a repeatable set of steps, they can be based on hard data and/or soft data and not all applicable to every race or horse, and some can alter according to xyz criteria i.e. from very simple such as back top OR if trainer a, b or c. To a big decision tree containing a myriad of variables.

In my experience, even the ones that work to whatever degree don’t always stay that way for various reasons.

Losing runs can be hard and the lower your average strike rate the worse they can be, I’ve done the maths in excel and built a bets simulator that I can plug in parms such total bets, strike and avg win odds, and they both came up with the same results e.g.

For a 1000 bets
a 20% strike you can end up with a losing run of 30
a 30% strike you can end up with a losing run of 19
a 40% strike you can end up with a losing run of 13

The clustering of wins and losses over an extended sequence can be very interesting i.e. whether backing doubles, trebles etc would have any mileage.

From what I've seen an incorrect/bad staking plan can turn a winning selection process into a losing one very easily.
 
Last edited:
Back
Top