Sharpe > 3 usually trades a lot. How do you control the costs of trading? Are you an exchange member? If you have such a high sharpe, you should join a hedge fund and make millions.
I see why sharpe of 3 is so desirable
- Thread starter SabreMan
- Start date
Sharpe > 3 usually trades a lot. How do you control the costs of trading? Are you an exchange member? If you have such a high sharpe, you should join a hedge fund and make millions.
The dradown of 3.74% seems very, very small and recovery of 3 days seems almost impossibly short for the standard deviation of 0.98% and mean of 0.2%. Your returns must have some kind of positive skewed distribution (not necessarily a bad thing at all
) did you make most of your money in a few big up days? Would be interesting to see the equity curve.
I meant changing the systems signals to opposite when the equity curve turns south.
You don't need to trade that much to get sharpe over 3. I think you are overestimating it. To get a sharpe of 15+, yeah, you need to trade a lot for that (like HFT).
The main thing I do to control costs is just to incorporate them into my models, so that I'm not doing any trades that are negative expectation after costs. This stops my system from trading lots of penny stocks where the trading costs are relatively high. Stocks with high price are super cheap to trade on a per-share schedule.
The other thing is being able to do enough volume to get discounts. At IB I'm doing enough trades to get into tier3 (0.0015/share) for a lot of my trading which I think is fairly decent.
Don't know anything about hedge funds as I'm not from a finance background and have no connections to that world. Exchange membership as far as I understand is for broker dealers which are large firms with at least dozens of employees and tens of millions of assets, not something available to regular traders due to the huge regulatory overhead.
Ah I see. Nah that isn't something that I typically do since in my experience if an edge stops working it tends to just go flat instead of totally inverting its behavior, and there's a chance it can start working again.
The dradown of 3.74% seems very, very small and recovery of 3 days seems almost impossibly short for the standard deviation of 0.98% and mean of 0.2%. Your returns must have some kind of positive skewed distribution (not necessarily a bad thing at all
Below reflects 13 months (june is the additional month). Notice the first half until about December my returns are less. I was trading half size still getting used to the system. I traded Full size starting in December and my confidence and execution of the system increased and then it exploded UP. (green line is SPX) I place 30 round trips per month roughly average hold let's say 6 hours or so. Max loss per trade is about 1.8% I've increased trade size 3 times in 13 months.
Here are the updated statistics, June has been a rough month for me but I'm at new highs, only up about 3.5%.
Last edited:
I see, so the distribution changes quite a bit between the two time periods, that explains most of what didn't make sense to me. Do you think your system also benefits from higher volatility? Regardless, very nice curve, congratulations on your success.
First half, I was dipping my feet in the water (trading half of projected size). Second half I traded full size, and was simply executing the system much better.
It does perform better in higher volatility periods (I believe). Basic reason is the moves in all most of the markets I trade have simply larger moves. But this is not a sell options game or futures spreads.... this is simply placing a bet ie Long Crude, looking to take down 0.90 cents on crude with a fairly wide stop. I'm only correct about 40 to 55% of the time. But 25% of the time I have scratch trades and the other 25% are losses.
Being that we have had relatively low volatility, this is something I need to learn: adjusting position sizing during perceived higher vol time frames.
For instance right now I am trading NQ's, yesterday and now.... I'm trading half size.
It does perform better in higher volatility periods (I believe). Basic reason is the moves in all most of the markets I trade have simply larger moves. But this is not a sell options game or futures spreads.... this is simply placing a bet ie Long Crude, looking to take down 0.90 cents on crude with a fairly wide stop. I'm only correct about 40 to 55% of the time. But 25% of the time I have scratch trades and the other 25% are losses.
Being that we have had relatively low volatility, this is something I need to learn: adjusting position sizing during perceived higher vol time frames.
For instance right now I am trading NQ's, yesterday and now.... I'm trading half size.
I could never get the need of 1M account with the 250K at risk.Why the extra 750K?
How many variables do you need to optimize for sharpe ratio of 4? Are you then set for life or do these system have a short shelf life?