Backtesting irregularities
I've been testing some strategies as of late with not that much success. I did start to notice however that many strategies that I try do well in Jan and Feb, but then take a dip in March and April. The same thing's been happening to different strategies that I'm testing. Im struggling to figure out why this is the case. I can understand this occurring for one or two similar strategies, but why would this be affecting different strategies that are tested under differing circumstances? I've tried to see if this is due to different race types but that doesnt seem to be the case. To add to the confusion, a bot that I have been running throughout all those months appears to be doing roughly the same. Has anyone experienced similar or can shed some light on this? Would love to hear your thoughts.
Yes it is indeed racing. I figured this may have something to do with it. I've read on the blog that this usually means that steams/drifts tend to run for longer - in theory this should help some of the strategies that im testing along, or at least not have a negative impact.
Is there someway I can account for this? ie limit some races as opposed to others?
Still finding more of the same though can't figure out whats so different about these months. Strategies that work in Jan/Feb dont in March/April/May and viceversa. Would someone please enlighten me?
Does anyone apply different strategies to different racing seasons? Whats makes them so different? I've read the blogs but to be honest posts don't shed much light. I'm a bit new to all this and have had some moderate success, but have been stuck on this issue for a while now. Any wisdom very appreciated
Does anyone apply different strategies to different racing seasons? Whats makes them so different? I've read the blogs but to be honest posts don't shed much light. I'm a bit new to all this and have had some moderate success, but have been stuck on this issue for a while now. Any wisdom very appreciated
-
- Posts: 3140
- Joined: Sun Jan 31, 2010 8:06 pm
How many seasons has the 'irregularity' occurred for? From the limited info you've given its hard to say especially if your bot is plodding along as normal. Are you sure it's not due to how you're approaching the markets during that period when there's few meetings, trying too hard etc
I never perform well in Jan/Feb but reckon that's down to too few meetings ,trying too hard and getting bored with a couple of meetings a day, no night racing etc
I never perform well in Jan/Feb but reckon that's down to too few meetings ,trying too hard and getting bored with a couple of meetings a day, no night racing etc
think about what happens when the flat season kicks in that doesn’t happen leading into the jumps season, I don’t have a different method as such from one season to another, with experience you’ll get to know the subtle differences that occur from season to season.sa7med wrote: ↑Mon Jul 02, 2018 5:47 pmStill finding more of the same though can't figure out whats so different about these months. Strategies that work in Jan/Feb dont in March/April/May and viceversa. Would someone please enlighten me?
Does anyone apply different strategies to different racing seasons? Whats makes them so different? I've read the blogs but to be honest posts don't shed much light. I'm a bit new to all this and have had some moderate success, but have been stuck on this issue for a while now. Any wisdom very appreciated
I don't do any manual trading so it's not a matter of trying too hard. I've had one bot working since Jan, and have been expanding it bit by bit but it has reached its limit now. So I've been looking into new strategies - mostly in the 5 mins pre-off but also in other areas.spreadbetting wrote: ↑Mon Jul 02, 2018 6:02 pmHow many seasons has the 'irregularity' occurred for? From the limited info you've given its hard to say especially if your bot is plodding along as normal. Are you sure it's not due to how you're approaching the markets during that period when there's few meetings, trying too hard etc
I never perform well in Jan/Feb but reckon that's down to too few meetings ,trying too hard and getting bored with a couple of meetings a day, no night racing etc
I only have data since this January so its just been tests on data since then, so yes, not much data to go by but whats got me confused is that im seeing this occur with many different tests of different strategies all together. So it's suggesting to me that there's something significant that I'm missing (despite the small data set). I realize that we're in the flat racing season now and that might well be the cause of these discrepancies - I'd just like to understand how/why they are the cause.
Yes, unfortunately I don't know that much about racing and have been working at this from a purely "markets" perspective so if you have any reading material to suggest I'd really appreciate itKorattt wrote: ↑Mon Jul 02, 2018 6:06 pmthink about what happens when the flat season kicks in that doesn’t happen leading into the jumps season, I don’t have a different method as such from one season to another, with experience you’ll get to know the subtle differences that occur from season to season.sa7med wrote: ↑Mon Jul 02, 2018 5:47 pmStill finding more of the same though can't figure out whats so different about these months. Strategies that work in Jan/Feb dont in March/April/May and viceversa. Would someone please enlighten me?
Does anyone apply different strategies to different racing seasons? Whats makes them so different? I've read the blogs but to be honest posts don't shed much light. I'm a bit new to all this and have had some moderate success, but have been stuck on this issue for a while now. Any wisdom very appreciated
- ruthlessimon
- Posts: 2094
- Joined: Wed Mar 23, 2016 3:54 pm
Annoyingly I don't have proper data for Jan/Feb
How would you class your strategies?
1. Momentum: Breakouts, pullbacks etc
2. Mean reversion: Fading moves etc
I've tried a diverse amount of things mostly in the realm of #1 but also lots of different scalping strategies under different criteria (mins pre-off and hours pre-off).ruthlessimon wrote: ↑Mon Jul 02, 2018 6:44 pmAnnoyingly I don't have proper data for Jan/Feb
How would you class your strategies?
1. Momentum: Breakouts, pullbacks etc
2. Mean reversion: Fading moves etc
At the moment, the only thing I can think of that could explain is that there's a marked increase in volatility on average post-feb - so therefore "false breakouts" and stop losses getting triggered more often - though I'm saying this with very little confidence
Feb 2018 was almost negative for me, lots of abandoned meetings, delays and crap liquidity so I wouldn't be afraid of ignoring that month.
I also find that during the transition months, jumps->flat->jumps to be a bit rubbish as well, I always assumed its due to the lack of data available for certain horses / breaks etc. so the smart money is not that smart.
I also find that during the transition months, jumps->flat->jumps to be a bit rubbish as well, I always assumed its due to the lack of data available for certain horses / breaks etc. so the smart money is not that smart.
are you on auto/manual or both? Part of whats got me confused is not that Jan/Feb are dull, but rather what works in these months tends to fail in March/April/May (usually getting worse towards May) and vice versa. I've actually back tested some things which come up with moderate returns in Feb but take a dip towards May. To compound the confusion, according to what I've read on here and on the blog about the transition to flat season, many of the strategies I've tested should do better in the flat season (if moves tend to last longer) but they perform worse. I think I need to take Mr. Webb's course. When's the next one??LinusP wrote: ↑Mon Jul 02, 2018 7:03 pmFeb 2018 was almost negative for me, lots of abandoned meetings, delays and crap liquidity so I wouldn't be afraid of ignoring that month.
I also find that during the transition months, jumps->flat->jumps to be a bit rubbish as well, I always assumed its due to the lack of data available for certain horses / breaks etc. so the smart money is not that smart.
not til October/ November I wouldn’t’ve thought, but not entirely sure it focuses on automation but may be wrong,
If you’re looking at going down the automation route it may be an idea to fully understand how the markets works manually before looking at automation?, just my opinion.