I don't understand

[volpri]

In short time frames where a high frequency of trades are taking place and where one is monitering actual risks vs initial risk and correlating those along with dynamic probability/dynamic reward vs initial probabilty/reward AND doing such things within a compounding modus operandi objective makes shorter term trading potentially much more profitable if a trader gets good at it.

The so called “noise” has much profit potential embedded therein...ROFL

 

[volpri]

But it does take time to learn how to do it and to make judgement calls on the fly and flexibility of mind to discard initial assumptions and fly with reality. Most will not be able to do it so for most it is probably best to use another paradigm of trading that better suits one’s abilities and tendencies because the so called “noise” can also chew up one’s account.

ROFLMAO

 

[IAS_LLC]

For as far as I know HFT's don't trade the ES. Problem might be the $12.50 fix prices per tick. But I am not sure about that.

And if they would, and I use your example, make $1,250 per contract per day, and trading 1,000 contracts (which is very small money for Virtu). They would make then 240 days*$1,250*1,000 contracts. Or in total 300 million$ a year. That's roughly twice the total yearly net profit of Virtu, trading 230 markets with 12,000 instruments by 148 people. Virtu should kick everybody out and just trade the ES then.

Something is not adding up correctly in your example. I will come back to this later, no time now.

PS: my postings were initially a reaction to OP. So not about Virtu, but about the 20 year old OP who wants to do HFT.

I was only making a point, which you seem to be missing. The point is individual trades don't matter...only the expected value of the trade, and how many times it can be replicated.

Also...

Your looking at their earnings, you need to look at their trading income. They have other costs outside of trading. Hopefully you don't trade fundamentals!

$180M of net adjusted trading income (not sure what adjusted means) for Q3 2018. They're doing $2.82M per day. Obviously, they're not only trading 1 instrument...and when your making markets, your expectancy isn't equal to the spreads...its less, but still positive and captured many times per day. 1000 contracts is only about 0.1% of the Daily ES volume....so them doing that is highly likely (if they make that market).

 

[Sprout]

Normally I would look at a 100 tick chart or less and scalp it for one or two ticks. Ideally I was trying to profit off of the spread, in simulated it works, but realistically there needs to a buyer for every seller and vice versa. I know the picture attached isn't considered scalping but its a system that has worked 80% of the time. So instead of capturing quick profits off of the spread, I am redesigning the algorithm to find volume surges from institutional algos to make quick profits. I primarily trade off of a 1 minute and 5 minute chart.

L means long
S means short
E means exit

my stops are 6-8 ticks back with 4-10 tick profit targets. Are my stops too tight?
to ensure execution I take liquidity 60% of the time. Dependent upon market conditions...

Having another T&S stream that filters out contracts <50 will give you more insight.

 

[volpri]

Try this. In a bullish surge let the surge happen where you see it on the chart. Immediately place a buy order 4 to 6 ticks back (away from the market). The odds favor that on a surge (with volatility) that price will drop back very quickly 4 to 6 ticks and sometimes up to 3 points before the surge quickly resumes again. Many times you can place the buy order away from the market like this and quickly get filled and seconds later be in a 4 to 10 tick profit.

This is scalping momentum but can be quite profitable especially doing it over and over and getting in each time (as much as possible) below the market. That reduces your risks and gives high probability but the downside is it tends to sacrifice the reward. You have go for a smaller profit but ...less risks..less time. The benefit is compounding profits.

See you never know how long a momentum surge will last in reversion instrument such as the ES. So you grab the profits when they are given to you. LOL

 

[schweiz]

I was only making a point, which you seem to be missing. The point is individual trades don't matter...only the expected value of the trade, and how many times it can be replicated.

I agree that individual trades don't matter. But the "expected value of the trade, and how many times it can be replicated" is only valid if reality is equal to what is expected.

I made a small excel to show you based on your example what the reality will be:

With 51% of winning trades and 1 tick profit per trade, only the green combinations are possible. So to make your 100 ticks in 100 trades you need 3 ticks profit per winning trade and only 1 tick loss per losing trade. HFT trading is about milliseconds, quickly in and out. 3 ticks in the ES is impossible in HFT as it will take far more than a few milliseconds, it can even take a few seconds. Virtu will never do that. Especially not 50 times within 1 day. And how can they work with a 1 tick stop???? All other possibilities are evenmore difficult as the winning trades would need to make at least 1 full point. In this model I don't even count any costs or commissions... So reality will even be worse.

Your looking at their earnings, you need to look at their trading income. They have other costs outside of trading. Hopefully you don't trade fundamentals!

My point was that all these expenses are a waste of money if they can make 300 million $ only trading the ES. And their expenses will be only a fraction as they don't need all these people just to trade the ES. So I compared correct: 300 million minus a bit of costs against XXX millions minus a huge cost that will result in half the profit of what the ES trading would generate alone.

Don't worry, I don't trade the fundamentals. And I also happen to have a degree in accountancy, so I know a little bit about balance sheets and P&L.

Since 2014 Virtu does not communicate the winning rates anymore, but at that time it was almost 51/49. After all the trades there was just a little bit of profit left. Average profit per share traded was then $0.0027. So your 1 tick per trade is at least very optimistic. Problem is also that they cannot trade unless they move the price by 1 tick as that is the smallest increment (unless they can cheat). For stocks that is completely different. They will never trade indexes with fixed price increments.

As I see your calculation I see you also misunderstood my 3 points. It was a 3 points stop, not a 3 points profit. Profits are much higher, so not $450. There are days (like today) that my profit is more then $1,250. Especially the last 2 months.

 

[IAS_LLC]

You're focusing on the wrong thing! Forget Virtu, forget 51%, forget ES, and forget 1 Tick stops.

You made the claim that risk is inversely proportional to trade frequency, which is completely false. Your also essentially claiming that longer term is better because you have a more favorable Risk to Reward ratio.... My point is all that matters is Expected Return on Trade (which is a function of win rate and win/loss size), and how frequently that trade can be replicated. At a certain point, assuming positive expectancy for both long and short term strategies, number of trades per period will trump risk to reward ratio and win rate.

https://en.wikipedia.org/wiki/Expected_value

Sounds like your doing great, and I'm not trying to change the way you trade. But this is just stochastic process math, and not really debatable.

If reality isn't representative of what is expected, there has been a structural break-down/change to the markets (which is why we have risk management) or worse... your expectation was based upon a study that wasn't statistically significant in the first place.

 

[schweiz]

English is not my first language so probably I don't fully understand you, or did not clearly explain what I meant.

You made the claim that risk is inversely proportional to trade frequency, which is completely false.

I wrote:"Risk is not related 1:1 with length of time." Can you tell me where I wrote that risk is inversely proportional to trade frequency? I don't find it. I found this:

"High" frequency strategies can have much higher sharpe ratios than longer period strategies have.

But that was a reaction on somebody else (gaussian), not me.

I will try to rephrase what I wanted to tell:
I tested my trading system in different timeframes. This lead to the conclusion that, starting from big timeframes, reducing the timeframe increased the profitability of the system. But at a certain point reducing the timeframe decreased the profitability of the system again. So I found an optimal point for my system where my performance was optimal. In this optimal timeframe I made less trades then in the very small timeframes, but made more money.
When I went to very small timeframes, profits per trade became smaller and slippage and commissions became high percentagewise. Going to very big timeframes would also generate smaller profits as reaction time (lag) on marketchanges become too long.

 

[murray t turtle]

I recorded my screen and slowed it down and figured out what is going... when the algo is placing a bid it gets filled immediately, even though the offer is not at that specific offer. So it's an order that wouldn't occur in the live environment. I'm working on fixing the issues as we speak.

%%
Something else helps. IS THAT BAD data?? NO not really beacuse SPY,[ AKA,S&P 500] NasdaQQQ can + have gone up more than 16% [monthly charts],OCT or NOV bear market.[Source my charts/Hirsch Stock Traders Almanac Easy to get bad data on smaller time frames.......

 

[trade4succes]

"risk is inversely proportional to trade frequency"

well, actually, if you do trades with the same (or scaled down) risk/reward, but you do more of them in the same time frame, then risk per time frame actually goes down obviously.