Crappest bot you'd run?

[eightbo]

I've got this old nothing burger strategy on the shelf which as u can see it's not doing much other than participating, kinda small sample size as it lets bets run but I was wondering:

comms_churn.png

What's the crappest edge strategy you would run?  maybe you have some storeis about your own choppy strategies that with a bit of filtering you managed to turn more + ? let us know let's discus$

got the CPU to resume it rn so gonna give this particular one another few months & if its still choppy/losing I'm gonna tighten up my conditions for an edge boost, but the tradeoff is losing 7/8 matched bets, so kinda takes a long time to get a decent return

also you can see I was desperate as I checked the weekday, has anyone on here found success by filtering by weekday or evening (more recreational punter money after work maybe)?

[Euler]

You should cut your strategies by weekday. I have a strategy that only runs on Saturday because the money mix is very different, so never rule anything out. Different grades of racing, different behaviour in the market. It's all valid as part of a strategy.

[ShaunWhite]

It looks promising on Monday and Tuesday, those days are characterised by being predominantly weak markets.

Rather than filtering by day, it might be worth trying splitting it by weak/strong. If the strategy does in fact work best on weak stuff, then you'd be able exclude the stronger ones on Monday Tuesday and include just the weak ones in the rest of the week. The name of the day itself isn't relevent, it's something about the market conditions on those days which is. Could be volume, volatility, race quality or something else, and whatever it is, those conditions will exist to a greater or lesser extent on any day of the year.

Tbh a strategy that works well on weak markets (where punters have little choice to take poor prices) is gold dust because most markets are pretty poor. And when you have a weak market strategy, the chances of it working on anything weak are pretty good, maybe dogs or aus horses. You won't necessarily be able to scale your stakes much, but you'll have a shed load of opportunities, and the more opportunities you have, the faster your intel accumulates.

[eightbo]

some good ideas, ty
it's back in live today, and I'm filtering grade (based on market name) in my results data to split into weak/strong as well

gonna watch it a bit first + try make informed decision before making a change

[firlandsfarm]

It looks promising on Monday and Tuesday, those days are characterised by being predominantly weak markets.

Rather than filtering by day, it might be worth trying splitting it by weak/strong. If the strategy does in fact work best on weak stuff, then you'd be able exclude the stronger ones on Monday Tuesday and include just the weak ones in the rest of the week. The name of the day itself isn't relevent, it's something about the market conditions on those days which is. Could be volume, volatility, race quality or something else, and whatever it is, those conditions will exist to a greater or lesser extent on any day of the year.

Tbh a strategy that works well on weak markets (where punters have little choice to take poor prices) is gold dust because most markets are pretty poor. And when you have a weak market strategy, the chances of it working on anything weak are pretty good, maybe dogs or aus horses. You won't necessarily be able to scale your stakes much, but you'll have a shed load of opportunities, and the more opportunities you have, the faster your intel accumulates.

How are you defining a weak market Shaun? If the book is as close to 100% as you can get then for each weak price there is a corresponding strong price that is value (if backing).

[firlandsfarm]

You should cut your strategies by weekday. I have a strategy that only runs on Saturday because the money mix is very different, so never rule anything out. Different grades of racing, different behaviour in the market. It's all valid as part of a strategy.

I've always read that if there is a distortion like in eightbo's stats, you should ask yourself "why?" because it could be a statistical anomaly. Maybe there is a genuine edge or maybe it's that blasted anomaly that will revert to mean as soon as you start following it with real money! I can see Saturday as being different to any other day and that on Monday and Tuesday you get poorer quality races ... maybe that's the Why?