Managing exits in manual trading

[Ghost_of_Blotto]

On another thread, garachen wrote

"I think the hardest thing about manual trading is managing exits. I find that, more often then not, peoples entries have positive expectancy but then they become way too fixated on trying to exit based on some metric related to entry price rather than then disappearance of alpha. "

It would seem logical that the entry price should be ignored as not being predictive of likely future price / liquidity at price. Trades which are not working as expected need to be cut as soon and as cheaply as possible. (based on an assessment of the future expected return being negative rather than price in relation to entry price)

If we remove rudimentary (and flawed) trade management / risk management strategies often discussed amongst manual traders (stop losses, whether dollar based or ATR/points based) - what other options have people explored for managing risk and managing trades?

For the purposes of this discussion I'm assuming a strategy which produces reliable entry signals for both the likely direction and magnitude of the expected move. However, the trader is going to need to use discretion/intuition to exit the trade as while there are criteria for expected move and taking a profit the trader will have to determine when the thesis for trade entry is no longer valid.

Especially interested in techniques to train a manual trader to develop intuition in trade management in order not to sabotage +EV entries while appropriately managing risk.

 

[Ghost_of_Blotto]

Here is an example of the fixation garachen referred to: here we have a manual trader who is risk averse and doesn't like positions to start going offside and who is training on a new product.

Two strategies used in this example

Strategy A seeks to capture an expected spike with limit order. Trader discretion used to assess when the likelihood of filling has diminished and the strategy will cross the market to exit. A competent operator will yield a profit (target was 14 ticks in this case) around 70% of the time with the remaining 30% being split between washes and a few ticks either profit or loss. Fees are about 1/4 of a tick. #1 is a trade under this strategy.

Strategy B trades when the probability of a significant (e.g. in this example 40-80 ticks) reversal of the prior direction is sufficiently high. Trades #2-4 are examples.

This operator is risk adverse and aims to be fairly precise in entry areas - works well for strategy A (high probability of small profit low chance of trade going offside) but not for strategy B (medium probability of very large profit but high chance of trade going temporarily offside potentially as much as around 1/3 of expected profit).

Chart attached as a visual aid - non chart method to give indication of areas to look to enter and the size of expected move. Green markers for long fills, red markers for short, white markers indicate going flat. Its down to the discretion of trader to time entry in these areas (using e.g. depth of market / time and sales information).

The trader has destroyed expected value using exit criteria which are not suitable for strategy B. However the entries are timed in the main to be in the bottom (top) 20% of the move for buys (sells).

Training seems worthwhile as this strategy would be very difficult to automate and the accuracy in predicting larger moves relative to distance from bottom or top where the reversal entry is signalled suggests that the method is highly scalable and could be applied across multiple similar products - if traders could be trained to manage the signalled positions profitably.

--and a little more detail on the executions:

#1 failed to fill at the expected area and market was crossed to exit for 2 ticks profit. No complaints. (strategy A)

the following are all strategy B, so they expect a large reversal.

#2. trader takes what is initially determined to be the appropriate entry into the signalled 60 tick upside but pitches the position when it doesn't get onside immediately. Failure to either re-establish position at a better price or hold.

#3. again, trader is exiting the position as its gone momentarily offside. position reestablished at a worse price and exit taken for a profit. These decisions resulted in -8 then +14 ticks, when the main strategy signalled 1 short which should have been profitable for at least 14 ticks.

#4. buy was signalled, but trader felt based on microstructure that it would be possible to re-establish at a better price. This was achieved in the second execution but again exited when the trade didn't get immediately onside.

For sure this is a pitfall of internalising the once useful advice to "get paid on a limit and cut the trades which go offside, run the profits for a few ticks" which was a pretty simple strategy to earn a tick per contract traded in thick and slow moving bond futures contracts in the early part of last decade.

 

[Redneck]

I have yet experienced trading consistency when applying discretion / intuition

However…, and last I checked – successful trading is certainly about consistency (or dumbass luck..., then quitting while still ahead - but where's the fun in that)

Quite the dilemma

=============

So…, me think it prudent we take a lesson from gangs…, by following their mantra of..., blood in – blood out

Which…, for the genteel / squeamish/ dolts among us…, equates to – rules in…, rules out


Though I’m cert there traders who (and repeatedly) find their self in trades – where bloodletting a much preferred endeavor


RN

 

[redbaron1981]

Maybe this thread is a little dated but it's a very important topic.

Having read a lot of Rearden Metal posts the way he enters trades quite often leads to negative posts from other forum members as he is using non coventianal risk management. One of his quotes sums this up.


"Here's something that even decade+ veteran traders have trouble understanding:

The market does not know or care where you got in!

Flame away, but all the best equities daytraders I know will ALMOST NEVER USE STOPS. 'Being up' or 'being down' on a position shouldn't ever change the way you trade it. The market has no idea where you got in. If you're long and the stock is slowly inching up during a general market selloff- would you cash out just because you're up? That's not a winning way of thinking. If you have a good feel for the market, trade based on what the stock is doing, never based on your entry."

Maybe one way to do this is to start off with a small position and increase into full size as the move goes into our favour. We can still have an absolute dollar value stop, but this value can give the trade enough breathing space not to get into a panic of a slightly red PnL and start making irrational decisions.

 

[ras72]

... Rearden Metal posts the way he enters trades quite often leads to negative posts from other forum members as he is using non coventianal risk management. One of his quotes sums this up.

"Here's something that even decade+ veteran traders have trouble understanding:

The market does not know or care where you got in!
Flame away, but all the best equities daytraders I know will ALMOST NEVER USE STOPS. 'Being up' or 'being down' on a position shouldn't ever change the way you trade it. The market has no idea where you got in. If you're long and the stock is slowly inching up during a general market selloff- would you cash out just because you're up? That's not a winning way of thinking. If you have a good feel for the market, trade based on what the stock is doing, never based on your entry."

Maybe one way to do this is to start off with a small position and increase into full size as the move goes into our favour. We can still have an absolute dollar value stop, but this value can give the trade enough breathing space not to get into a panic of a slightly red PnL and start making irrational decisions.

In general there's something intriguing about the “(mostly) no stops” tenet.

Specifically, in my experience of losing trader, my longest successful run of three months was when trading without stops and also averaging (there's another provocation!), building around a small position, accumulating, reducing, reversing, in a very dynamic, fluid process. Playing FX, 400:1, daytrading, discretionary. Failure came due to emotion, psycho loss of focus, analysis confusion...

Difficult to evaluate separate from other trading components. As with any other betting strategy it won't overturn game expectancy; still required to be prevalently correct in directional calls.
Liberating experience. Can give (obviously) unjustified sense of invulnerability; can overextend and crash (...yeah, some kind of hard stop may be necessary). A necessity for some due to size. A preference for others.

Still believe in it. I wonder if one should simply ignore exits and only trade entries. Stops and mechanicals cramp my style.

Somebody on here noted that: “A Stop Loss is really a Take Loss and a Take Profit is really a Stop Profit”