There’s something I was wondering about. On the USD/JPY (H4) chart we have a extreme low at 76, so actually we are still in an uptrend. But is there any rule how to connect the up trendline with the second point? Take a look at this chart:
If price moves again up we would have a good up trendline because there seems to be some support. But if it moves down, does it mean there is actually no real uptrend? Because I could have connected it on every 4 hour candle in the last few days and it would move again below the line.
An uptrend you say? I would say, what you are calling an uptrend was / is a retrace, where the downtrend may continue. Look at the daily and weekly chart, price has yet to breach the long term down trend since 2007.
There was a reason the UJ dropped down to the 76 level and even with the disaster in japan, and with the USD still tumbling there is no reason to think that the down trend is over just yet.
My chart is too big to post and don’t have to time to fool around with it at the moment.
Hi, sorry that I am late to this - I have been on holiday so have had ten days without trading! - but hopefully not too late to have something relevant to add. Two things occurred to me from the first post here, which I thought it worth sharing. I am largely a trend-based trader, and for me I need three touches before I will trust a trendline. That first chart posted simply showed the second touch at the low at 76.40 or thereabouts, so for me that is not yet strong enough to be considered a trendline. Secondly, as a previous poster stated, that low was the result of a very unusual spike. I find it much more reliable to consider only ‘normal’ sized bars when looking to draw a trendline. Trendlines are subjective, certainly, but they should still reflect the overall trend. A one-off spike is generally a reflection of an immediate reaction, rather than a reflection of how Price is moving over a sustained period. Often, a news announcement or such will cause a spike or dip in Price, then trend will revert to how it was previously. I like to use an ema on my chart to reinforce my view on a trendline, effectively using it as checks and balances to keep myself honest, I find it more reliable than simply entering off a trendline that I just put on the chart. The ema for that Yen period is nowhere near the trendline, which would always lead me to question a trendline in any case.
Anyway, apologies, I am rambling. But my point boils down to this: when drawing trendlines, I look for more than two touches before I consider it a trend, and I would not use spikes such as that one to form a touch, I find that ‘normal’ bars work much better.
Incidentally, for me, USD/JPY has been in a nice downtrend on the Hourly chart, supported by the 240, for most of April. The Daily had been showing a downtrend for much of the latter part of last year, with many good trading opportunities, then broadly ranging this year except for the mid-March movements which are only now clearing fully. I have not seen an uptrend on the Daily that I would like to trade in some time.
No not rambling ST, you reinforce a very good point.
There is a subjective element to trendlines, but I keep thinking that’s less and less true , there is the element to which purpose do you use them for too.
Good point about 3 point for a trendline, but just food for thought - here’s a valid 2 point trendline, I may be looking at a long at the 4650 support line, so if the price meets where the two lines converge, it could be a goer.
Thank you! I see where you are coming from, but for me to consider that a trendline I would look for Price to give me the third touch at the bottom of the current move - so essentially I would view that chart as an unconfirmed trendline, waiting on the third touch. I don’t trade the 15 (tend to go for either the 5 or the 60, nothing against the 15 I just only have so much capacity for juggling TFs!), but that could be a downward channel. However, if you add some emas on the 15 (I just tried it), you could say that Price on the 15 has been broadly respecting the 21ema in an uptrend since yesterday morning, then this morning broke through that and the 50 but has found support at the 100ema - if that is the case then there would be a bounce at the 1.4650 level you mention. So I think that we are agreeing, basically!
To me it looks as though Price will either break through that level, confirming your trendline, or will find resistance there and reverse, giving the opportunity to go Long at 1.4650ish and labelling the current fall as a retracement.
Another issue with trendlines is that the wick on a candlestick chart can sometimes breach the line while not signalling the invalidation of the trendline… but that is the really subjective end of this area!!
Either way, 1.4650 will be interesting to watch for sure.
Well it didn’t hold!, but just to show you what I was on about
From point 1, the price is going down and so is the volume, that indicates no interest to go down, then at point 2 there is high volume which shows there is interest but where’s the price going? So it’s off to point 3 where we have the same thing, high volume but no further movement in price, then notice the very last candle with high volume breaking through resistance which suggest to me that the market has gone bearish, but wait and see as always.
One thing I recently learned about trendlines is that they are very geometric in nature. I honestly don’t understand this phenomenon, because I would’ve thought the market was a lot more “random” than that. However, knowing that price moves geometrically, you can not-only draw current and previous trendlines, but also projection lines. I would stake money on price stopping, at least briefly, at nearly every single geometric boundary. And the greatest thing of all is that there seems to be an infinite number of geometric patterns. This also corresponds to my belief that everyone can prosper at trading.
Your original question was about drawing a trendline connected to a specific bar. I suggest you go ahead and pick a bar … pick any one, or perhaps pick one that looks more significant. Then find the geometry of that trendline vs. others on the same chart. You might be amazed at how much information this tells you
