For us Shaun it's every hr after 1amShaunWhite wrote: ↑Mon Jan 08, 2018 6:38 pm(btw when i said "3000 hrs" above, i actually meant "every waking hour")
Strategy Development: Modelling
- ruthlessimon
- Posts: 2094
- Joined: Wed Mar 23, 2016 3:54 pm
- ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
My sleep's shocking atm. A log would look like a perm 5 from 24 matrix.ruthlessimon wrote: ↑Mon Jan 08, 2018 6:39 pmFor us Shaun it's every hr after 1amShaunWhite wrote: ↑Mon Jan 08, 2018 6:38 pm(btw when i said "3000 hrs" above, i actually meant "every waking hour")
Try trading at random. When I first started trading I did that and hardly lost anything. So I then developed a strategy that didn't trade at random and noted where the big losses came from. I then started working on striking them out.
Late to the party
Peter two questions if I may:
(1) What do you actually mean by trading at random? Seems a silly question but I assume you mean come up with a basic idea and then apply it in random markets?
(2) Typically how many trades do you think is statistically valid "per evolution" to make judgements and start nudging in a particular direction? This would seem fundamental to have a high enough confidence factor that those big losses were themselves not random and indeed generated by the system parameters.
- ruthlessimon
- Posts: 2094
- Joined: Wed Mar 23, 2016 3:54 pm
Really like the question, as I've been trying to answer a similar "thought experiment". One which Shaun also alluded to in a previous postCards37 wrote: ↑Sat Mar 10, 2018 11:36 pm(2) Typically how many trades do you think is statistically valid "per evolution" to make judgements and start nudging in a particular direction? This would seem fundamental to have a high enough confidence factor that those big losses were themselves not random and indeed generated by the system parameters.
When it's valid to say, "I've learned something new.", "this edge no longer works" etc
Let's say we have a day of 10 straight losing markets: Is that simply the expected probability of a longterm edge? (i.e. edge shouldn't be changed, nothing was learned, this was expected) Or was there a bias on the day? (i.e. edge needs to be adjusted to incorporate this new info). How do we tell the difference? Analysing every turn of the market, & you risk never seeing the bigger picture. At the same time, a huge dataset, will be extremely slow adjusting to genuine changes in the market. There has to be a balance
I am no statistician, but doing a Chi Squared test may provide the answer. It's not something I'd be obsessing about though.
Providing your betting bank is an amount you can cheerfully afford to lose, I'd say the rational thing is just to carry on through bad periods.
Worst case scenario = You lose your bank or have a huge drawdown
Best case scenario - You come out of the drawdown and make decent money.
People seek certainty, but like the blue in the sky, it's an illusion. If you want a guarantee, buy a toaster.
J
Providing your betting bank is an amount you can cheerfully afford to lose, I'd say the rational thing is just to carry on through bad periods.
Worst case scenario = You lose your bank or have a huge drawdown
Best case scenario - You come out of the drawdown and make decent money.
People seek certainty, but like the blue in the sky, it's an illusion. If you want a guarantee, buy a toaster.
J
ruthlessimon wrote: ↑Sun Mar 11, 2018 5:41 pm
Let's say we have a day of 10 straight losing markets: Is that simply the expected probability of a longterm edge? (i.e. edge shouldn't be changed, nothing was learned, this was expected) Or was there a bias on the day? (i.e. edge needs to be adjusted to incorporate this new info). How do we tell the difference? Analysing every turn of the market, & you risk never seeing the bigger picture. At the same time, a huge dataset, will be extremely slow adjusting to genuine changes in the market. There has to be a balance
What was the next step, after you'd eliminated the big losses?
Jeff
Jeff
- ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
Would it be fair to characterise your process as...
1. Devise strategy
2. Test strategy
3. Identify worst losses
4. Alter strategy
5. Goto 2
And somewhere in the loop you either discard it or save it as fully optimised.
To be able to see why some of your scalps made big losses, did you have more than the results to look at, ie were you able to see the precise market conditions around the times it was failing? Or were you eliminating on the basis of the fixed attributes, race type, class etc.
1. Devise strategy
2. Test strategy
3. Identify worst losses
4. Alter strategy
5. Goto 2
And somewhere in the loop you either discard it or save it as fully optimised.
To be able to see why some of your scalps made big losses, did you have more than the results to look at, ie were you able to see the precise market conditions around the times it was failing? Or were you eliminating on the basis of the fixed attributes, race type, class etc.
- ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
There's an old Bing Crosby tune that goes....
You've got to accentuate the positive
Eliminate the negative
Latch on to the affirmative
Don't mess with Mister In-Between
Seems appropriate.
You've got to accentuate the positive
Eliminate the negative
Latch on to the affirmative
Don't mess with Mister In-Between
Seems appropriate.
- ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
I see Mr Wing-It is hanging around in the strategy development thread No disrespect but I didn't think this topic was your scene. No need to reply to that, I'm trying not to push my question to Euler too far back.
There is a balance to be had. I figured right at the very start of my career I could pick off opportunities, it was the losses that were a problem.
I then started to flip and flop strategies depending on what I thought should work. You can cut and refine entries to the point where you don't make any. So it's always a balance of risk and reward.
You don't have to trade every market for sure and you can pick odd the odd one or two, but finding a way of profiting from most is how you take trading to the next level.
I think that’s a pretty important point. I’ve found the best strategies have the simplest principles behind them. As soon as you start tweaking, fiddling, filtering, you’re more likely to end up with something that either places very few bets or just doesn’t work at all because you’ve overfitted to past data.
Have a good reason why something should have a small edge (above random trading) that’s applicable in as many runners or situations as possible. Backtest the idea, and if it doesn’t work, move on to the next idea.
I recently found an edge that’s so simple I could describe the entire strategy (with parameters) in a sentence. That’s not to say it was easy to find. I might go months and months without finding anything new and profitable, but often when I do it’s quite simple.