You are 100% Certain to lose 99% of the time 1/2 the time.
- Thread starter TradingLogic
- Start date
I also don't agree entirely that leverage helps. It just increases your return, and your vol. Data fees and other expenses aren't that high. So most of your costs aren't fixed costs; they'll scale with size. If your return is negative after commissions leverage will just make it more negative. However if you can do better than breakeven then of course leverage can give you a decent return; at the cost of more risk.
It's for that reason I prefer to work out costs in sharpe ratio units. Actually the cost of SPY isn't as bad as you say - its one of the cheapest things we can trade (due to a combination of high price, tight spreads and high vol).
Commission on 200 shares with IB is $1, or $2 per round trip
I'd say slippage is about 0.02 per share (bid ask is 0.01), on 200 shares $4
So on 200 share blocks we're talking $6 a trade.
Vol is about $32 a year, or $6,400 per block.
Costs then are $6 / $6400 = 0.0009375 SR units.
Doing 250 round trips a year comes in at 0.23 Sharpe Ratio.
In comparision the NASDAQ future costs around 0.18, and this is one of the cheapest futures I know. If we can't make day trading work on these two things we can't make them work at all.
What proportion of your account that is; well if you were unleveraged and buying $200 x 200 shares each time you'd have a $40,000 account and it would be costing you $6 x 250 = $1,500 = 3.8%. Data costs might be what $120 a year? Lets keep them out of it to keep things simple.
If we double our leverage and buy 400 shares with $40000 costs exactly double (at this level of trading fixed costs aren't a problem) to $8 a block, or 7.6% a year. There might also be another 1% or so to pay in margin costs.
Alternatively if we halve our account size and keep the same to double leverage again it will still be 7.6% a year, or maybe 8.6% a year with margin (though fixed costs will go up slightly).
The point is with no leverage or double leverage if you can't make 0.23 SR then you'll lose money. If you make 0.20 SR then you'll make pre costs 3.2% a year unleveraged, which means you'll lose 0.6% a year. If you double the leverage you'll make 6.4% a year unleveraged, and lose 1.2% a year, or maybe 2.2% a year with the margin on top.
Now making 0.23 SR on one instrument is I think reasonably easy. I reckon I can make 0.40 SR for example. However I'm personally uncomfortable with giving up more than half my return in costs; receiving a net small sharpe ratio, and having to borrow big to make okay money. Unleveraged I'd only make 2.7% a year; to get to my current expected average return of 18% a year I'd need to leverage like eight times - far too much. Since with the stuff I trade I can get 0.40 SR trading daily, or once a month, there is no benefit to day trading. Instead I expect to pay about 0.05 SR a year in costs.
But suppose we can make SR 0.80 or 12.8% a year gross. Then unleveraged we'll earn 9.6% a year after costs. With double leverage we'll make close to 20% (I wouldn't personally pump it up any more than that).
If you can make 0.8 SR pre cost and 0.57 post day trading, then these costs are pretty decent. This doesn't seem improbable (I've seen figures on this site implying sharpe ratios of 13..... which is a bit harder to swallow).
Of course the problem is it doesn't seem that many people can make SR 1.0 day trading one instrument, or perhaps they are all just trying to do it with more expensive stuff.
It's for that reason I prefer to work out costs in sharpe ratio units. Actually the cost of SPY isn't as bad as you say - its one of the cheapest things we can trade (due to a combination of high price, tight spreads and high vol).
Commission on 200 shares with IB is $1, or $2 per round trip
I'd say slippage is about 0.02 per share (bid ask is 0.01), on 200 shares $4
So on 200 share blocks we're talking $6 a trade.
Vol is about $32 a year, or $6,400 per block.
Costs then are $6 / $6400 = 0.0009375 SR units.
Doing 250 round trips a year comes in at 0.23 Sharpe Ratio.
In comparision the NASDAQ future costs around 0.18, and this is one of the cheapest futures I know. If we can't make day trading work on these two things we can't make them work at all.
What proportion of your account that is; well if you were unleveraged and buying $200 x 200 shares each time you'd have a $40,000 account and it would be costing you $6 x 250 = $1,500 = 3.8%. Data costs might be what $120 a year? Lets keep them out of it to keep things simple.
If we double our leverage and buy 400 shares with $40000 costs exactly double (at this level of trading fixed costs aren't a problem) to $8 a block, or 7.6% a year. There might also be another 1% or so to pay in margin costs.
Alternatively if we halve our account size and keep the same to double leverage again it will still be 7.6% a year, or maybe 8.6% a year with margin (though fixed costs will go up slightly).
The point is with no leverage or double leverage if you can't make 0.23 SR then you'll lose money. If you make 0.20 SR then you'll make pre costs 3.2% a year unleveraged, which means you'll lose 0.6% a year. If you double the leverage you'll make 6.4% a year unleveraged, and lose 1.2% a year, or maybe 2.2% a year with the margin on top.
Now making 0.23 SR on one instrument is I think reasonably easy. I reckon I can make 0.40 SR for example. However I'm personally uncomfortable with giving up more than half my return in costs; receiving a net small sharpe ratio, and having to borrow big to make okay money. Unleveraged I'd only make 2.7% a year; to get to my current expected average return of 18% a year I'd need to leverage like eight times - far too much. Since with the stuff I trade I can get 0.40 SR trading daily, or once a month, there is no benefit to day trading. Instead I expect to pay about 0.05 SR a year in costs.
But suppose we can make SR 0.80 or 12.8% a year gross. Then unleveraged we'll earn 9.6% a year after costs. With double leverage we'll make close to 20% (I wouldn't personally pump it up any more than that).
If you can make 0.8 SR pre cost and 0.57 post day trading, then these costs are pretty decent. This doesn't seem improbable (I've seen figures on this site implying sharpe ratios of 13..... which is a bit harder to swallow).
Of course the problem is it doesn't seem that many people can make SR 1.0 day trading one instrument, or perhaps they are all just trying to do it with more expensive stuff.
lol, that's awesome. a beautiful example of playing to lose, no matter how the target is reached.
So Surf proves that it is possible to be consistently profitable. He even shows with what kind of trading it will work.
Those who say that being consistently profitable is impossible look like idiots now. Even people trying to lose money make more money than they can. No wonder they say all the time that being profitable is impossible.
So this event will cause again a depression for lots of people who will see confirmation that they are losers.
Being consistently profitable is possible. However, you need to be able to change as the market changes--- if you don't change your methods as the market morphs, you start to see your system degrading until it fails to work.
A sad example of this is the scriblers thread where the marketdeluded, but well meaningfolks, have confused hindsight and charts with executing strategies in the real world. Everyone should take this as a warning of what can happen if you stay fixated on a static system. Surf
Last edited:
%%%%%%%%%%%%%%%%%%%%
Thanks for the post;
trading /investing is NOT gambling.IF you think it, is compare brokerage winnings- with black jack winnings-that is if the casino lets you out with them.LOL Its not a coin toss also
Wisdom is profitable to direct
It is unbelievable that we never agree but basically tell the same things. You write now: However, you need to be able to change as the market changes
I wrote this some time ago in another thread: A good system is "autoadaptive". It should first see what kind of market you are trading and adapt itself to take optimal profits. So no matter what market you are in you should always make profit on a short term basis (day or week, not trade by trade). There is no universal system to trade all markets at all times in all kind of trends (or no trends). So a system should exist of a number of subsystems.
It's all about number crunching. All markets have prices, these prices are the basis to work on. Whether you trade futures, forex or stocks should not make any difference, because a computer calculates always the same way with numbers. 1+1=2 no matter if it is a quote from futures, forex or stocks. It will always stay 2.
Just in case somebody would say I did not write this posting: http://www.elitetrader.com/et/index...ount-waste-of-time.289658/page-3#post-4087049 from 2 months ago.
Doesn't matter. I've lost count of how many times I've said that a proper price action approach is self-correcting and self-adapting since it by definition has to be. But one may as well be arguing evolution to a pentecostal.
I suggest you shrug your shoulders, heave a big sigh, and go on making your money. That is, after all, what counts, not scoring points on message boards with anonymous strangers.
I understand what you are saying. Intuitively you move your entry points based on price action. I get that. However, this only works in higher volatility regimes---