**** me... huge realisation about why trend followers survive longer...

DashRiprock said:
well following on from my post above, lets say that the two traders we're comparing:

Mean Reversioner makes decisions on 80% order flow 20% information
Trend Follower makes decisions on 20% order flow 80% information

so in their ratio's they are equal but opposite. they will deffo be trading at different frequencies.
Click to expand...
At the risk of over-simplification, an MR trader taking a view that an asset was over-priced and therefore had a higher probability of reverting to the mean than any other development, would trade it short to around the point where it met the new mean value (or beyond allowing for over-shoot if they were somewhat more aggressive). MR traders normally do have an underlying trend bias in that the mean is rarely flat and they will typically only trade the contra underlying leg. In this example, the underlying (the mean) is downward and they’ll trade the short from above the mean, but less likely to trade Long from beneath.

A Trend trader will (normally) be found trading only on the ‘correct’ side of the mean and in this example of a downward trending asset will likely be setting up for his entry as the MR chappie is preparing to exit his.

Even in an idealised scenario with an aggressive MR trader that will take both legs of price deviation from the mean, he’s still only going to take a maximum of two trades to our trending guy’s one. That is a significant enough basis for assessing potential longevity of the two types of trade and trader, but as stated, that is an idealised situation which is unlikely to hold in reality.

My view is that the MR trader may push a few more trades through than your typical trend player, but not a statistically significant number.
 
TheBramble said:
OK. Just for clarification, are you suggesting that it would be more likely that our average MR trader would likely be trading a shorter time frame than your average Trend trader?
Click to expand...

well, i think its very likely that the average time in trade is more for a trend follower than a MR trader... but that is more a consequence not causation of my point.

my point is that orders pass through much fore frequently than new fundamentals/information (its all information i know, but you get the gist).

so, for example, a MR trader might see a "standard" trade as fading a spike in the crack spread because it was put out by a big order going through in crude

whereas a trend follower might go long crude because the market thinks threat of military actions in libya (the actual signal might be technical based, but its the fundamentals thats moving it).

if you say each of these are "standard" MR and TF trades respectively, in that they both contain equivalent units of information, its obvious that the MR trader will see more "information units" per time than the TF trader.

see what i mean?
 
TheBramble said:
My view is that the MR trader may push a few more trades through than your typical trend player, but not a statistically significant number.
Click to expand...

I'd agree with that, albeit on a fairly limited sample size of, er, me.

My main strategy is fundamentals based, but uses TA to refine entries/exits. Once I've done the work that shows a stock is a long or a short, I'll look for an entry and so might end up either being trend following (eg a 'long' which is already doing well and has broken out of a resistance level), or mean reverting (eg a 'long' which is a weak performer and is coming off support). I've had a quick look at my trade log and can't see any real difference either in the performance of the two, or the frequency with which I trade them. Although, to be fair, if I have an 'edge' it will be in the fundamental analysis that selects the longs and shorts in the first place, rather than my TA.
 
TheBramble said:
At the risk of over-simplification, an MR trader taking a view that an asset was over-priced and therefore had a higher probability of reverting to the mean than any other development, would trade it short to around the point where it met the new mean value (or beyond allowing for over-shoot if they were somewhat more aggressive). MR traders normally do have an underlying trend bias in that the mean is rarely flat and they will typically only trade the contra underlying leg. In this example, the underlying (the mean) is downward and they’ll trade the short from above the mean, but less likely to trade Long from beneath.

A Trend trader will (normally) be found trading only on the ‘correct’ side of the mean and in this example of a downward trending asset will likely be setting up for his entry as the MR chappie is preparing to exit his.

Even in an idealised scenario with an aggressive MR trader that will take both legs of price deviation from the mean, he’s still only going to take a maximum of two trades to our trending guy’s one. That is a significant enough basis for assessing potential longevity of the two types of trade and trader, but as stated, that is an idealised situation which is unlikely to hold in reality.

My view is that the MR trader may push a few more trades through than your typical trend player, but not a statistically significant number.
Click to expand...

OK i see what you mean i think

but what youve said here, to me, is that the MR trader is trading the pullbacks while the TF trader is trading the breakouts. but this is with both traders using the same information just different strategies

what I am on about is that MR traders make trades from different information to the TF traders.

so i guess we could both be right or both be wrong :)
 
DashRiprock said:
see what i mean?
Click to expand...
I do. If I understand you correctly, you're seeing most trend traders operating in longer TFs than most MR traders. I have no idea how many in each camp operate in each TF, which goes back to my earlier email.

But to give an example. GBPJPY right now 12:47 GMT.

A trend trader on the M15 may be trading what he sees as an underlying up trend and the current pullback may be presenting him with a value entry opportunity. Just as soon as that 1 min starts to lift (or makes a HH or whatever) he'll enter. He's still the 'correct' side of his mean (200 or whatever) on his 15 min trading timeframe but he's effectively playing a MR trade back up toward that mean on the 1 min.

I'm not trying to blur distinctions, but point out in the shorter TFs (which may translate into the longer) your trading style is very much specific to that TF you are trading - and that TF alone.

As you move into the shorter TFs, where many trend traders do live, I don't think they will be executing a wildly different number of trades than their MR cousins.

Which is counter to Arabian's opening hypothesis.
 
TheBramble said:
At the risk of over-simplification, an MR trader taking a view that an asset was over-priced and therefore had a higher probability of reverting to the mean than any other development, would trade it short to around the point where it met the new mean value (or beyond allowing for over-shoot if they were somewhat more aggressive). MR traders normally do have an underlying trend bias in that the mean is rarely flat and they will typically only trade the contra underlying leg. In this example, the underlying (the mean) is downward and they’ll trade the short from above the mean, but less likely to trade Long from beneath.

A Trend trader will (normally) be found trading only on the ‘correct’ side of the mean and in this example of a downward trending asset will likely be setting up for his entry as the MR chappie is preparing to exit his.

Even in an idealised scenario with an aggressive MR trader that will take both legs of price deviation from the mean, he’s still only going to take a maximum of two trades to our trending guy’s one. That is a significant enough basis for assessing potential longevity of the two types of trade and trader, but as stated, that is an idealised situation which is unlikely to hold in reality.

My view is that the MR trader may push a few more trades through than your typical trend player, but not a statistically significant number.
Click to expand...

Brambs

I tend to agree with your final point but, as many attempts to "label" activity, there are so many variations on the themes. For example, you could argue that someone playing an upward channel had a foot in both MR and Trendy camps.

Similarly, in that same channel scenario, it would depend on the set-ups employed as to how many trades either camp would make. An MR would likely take 2 - long at bottom of channel followed by short at top - but a Trendy with both retracement and break-out set ups would also be taking 2 and so on.

I suppose if you designed some strict definitions of MR activity and Trendy activity you could determine some variation in number of trades likely - but would there be much point in doing so?

good trading

jon
 
TheBramble said:
I do. If I understand you correctly, you're seeing most trend traders operating in longer TFs than most MR traders. I have no idea how many in each camp operate in each TF, which goes back to my earlier email.

But to give an example. GBPJPY right now 12:47 GMT.

A trend trader on the M15 may be trading what he sees as an underlying up trend and the current pullback may be presenting him with a value entry opportunity. Just as soon as that 1 min starts to lift (or makes a HH or whatever) he'll enter. He's still the 'correct' side of his mean (200 or whatever) on his 15 min trading timeframe but he's effectively playing a MR trade back up toward that mean on the 1 min.

I'm not trying to blur distinctions, but point out in the shorter TFs (which may translate into the longer) your trading style is very much specific to that TF you are trading - and that TF alone.

As you move into the shorter TFs, where many trend traders do live, I don't think they will be executing a wildly different number of trades than their MR cousins.

Which is counter to Arabian's opening hypothesis.
Click to expand...

I think my post before this one answers some of your points but to eloborate a bit

if, say, you look at a 1m chart and a 15m chart and a daily and a weekly chart, some people say that you can't tell which one is which. I dont know if its true or not but lets run with it for easyness. in these situations, i think you are right and that a MR and a TF trader on the same graph will roughly make same number of trades (some people say that markets range more than they trend, but this is off a tangent). so i agree with you for all the traders trading just from graphs that time doesnt make a difference between number of MR and TF opportunities.

BUT what makes the price move on the 1m chart of waaay different to what makes the price move on the weekly chart. i guess my assumption is that for whatever type of trader you are (a MR or TF er) you know what is behind the trades you make (order flow or fundamentals) and so you look at the information that is appropriate for that.

so I mean its stupid to try and make a tick or 2 because of GDP forecasts just like its stupid to hold for weeks and weeks because there is a fake order in the offer and stops behind it.
 
barjon said:
I suppose if you designed some strict definitions of MR activity and Trendy activity you could determine some variation in number of trades likely - but would there be much point in doing so?
Click to expand...
Agree Jon. A definition of terms would be not only useful if we were really interested in the outcome, but it is vital.

As this was Arabian's thread, I suppose we should leave him to expend neceesary effort in stating more clearly his hypothesis.

I'm still not convinced he really meant to start a sensible thread anyway, so let's see how that one develops.
 
TheBramble said:
Agree Jon. A definition of terms would be not only useful if we were really interested in the outcome, but it is vital.

As this was Arabian's thread, I suppose we should leave him to expend neceesary effort in stating more clearly his hypothesis.

I'm still not convinced he really meant to start a sensible thread anyway, so let's see how that one develops.
Click to expand...

I don't see that that's my duty; threads are "owned" by the posters, not the starters... Dash is doing a perfectly acceptable job of saying what I meant :)

As for the other thing, I couldn't comment...
 
by the way from now on i think the traders that have absolutely no clue about what is behind the action they are trading (order flow, fundamentals, arbitrage or whatever) should be said to be "pulling a Howard'.

example:

"Did you hear about danny? he did his conkers doing long usd/jpy"
"really? shame. Why did he do that?"
"he said that because kerry katona was voted off dancing on ice it should go heavy bid"
"WTF? Dude... he pulled a Howard on that one".


LOLOLOLOL
 
DashRiprock said:
example:

"Did you hear about danny? he did his conkers doing long usd/jpy"
"really? shame. Why did he do that?"
"he said that because kerry katona was voted off dancing on ice it should go heavy bid"
"WTF? Dude... he pulled a Howard on that one".
Click to expand...
Erm, so that wouldn't be a good reason then?













s h i t
 
i think this was a good thread but the jokes have maybe spoiled it?

i mean jokes are good but there was proper trading talk for a bit about different views of me and bramble and shadowninja and robster970 and ect?

be a shame to not have good trading talk when its under our noses i think.
 
DashRiprock said:
i think this was a good thread but the jokes have maybe spoiled it?

i mean jokes are good but there was proper trading talk for a bit about different views of me and bramble and shadowninja and robster970 and ect?

be a shame to not have good trading talk when its under our noses i think.
Click to expand...

Wow. The arbiter of lulz. I would have never guessed.
 
You must log in or register to reply here.

Similar threads

V
Trend Following
2 3
Replies
42
Views
35K
wavebreak55
W
horace
Has anybody found trailing stops help trend following strategies?
2
Replies
28
Views
10K
JRP2891
J
F
t/a Market timing versus predicting roulette
2
Replies
37
Views
11K
tomorton
T
A
No Predictions for Trend Following
Replies
5
Views
2K
NVP
NVP
C
Trend Following
Replies
2
Views
2K
cube101
C
Top