Trading Analysis - 4th-9th Feb 2019


10 February 2019

As this week has been rather busy, and there is a clear suggestion of shorting behaviour why don't we do a little light trade analysis ?

Monday, 4th Feb 2019

 1.11M Buys vs 1.07M Sells - A degree of recovery towards 40p. Roughly 600K of >30K+ Buys and 630K of Sells - no obvious shorting sequences. Relatively normal trading behaviour and volume.

Tuesday, 5th Feb 2019

Tuesday is quite different - in the half hour after 8:48 there were apparently 13 sells over 30K (3x 50K and 4x market size 30K) and only a single Buy of 30K or over.  The rapidity of the trades and their specific sizing is highly suggestive of short selling, let alone the significant troll posting metrics at the time. This selling impacted the nett trade position by approximately 550,000 shares and we might conservatively ascribe 60% of this to short selling - thus 350K at an average price of 35.80. All 13 of the significant Sells are all highlighted below - you decide which sequences look particulary suspicious.

Buying in volume was regular and the nett MM trade position (note this is completely different to the amateur shorter's short position of 350K) rose from -550K to +300K at the end of trading  - i.e. more Buys (1.71M) than Sells (1.408M)

Wednesday, 6th Feb 2019

Wednesday the suspicious activity is not quite so concentrated, though the end result is very similar - a days end nett excess of Buys of c. 250K.

The first suspicious sequence starts around 9:07AM - we see a tight sequence of 3x 50K Sells which has the effect of dropping the Ask by 1.0p. Interesting there is also a strange sequence of trades reversed and then resubmitted at a new price - this could of course just be coincidence.

Following some large Sells - 92K, 240K, 107K which dropped the MMs nett trade position to minus 450K there was a second suspicious sequence  - nominally starting around 11:02 AM, but really kicking off around 11:19AM, we see a strong sequence of selling right at the Bid level of 35.0p. This sequences continues until the Bid is forced down to 34.50, at which point it appears to stop.

Again despite these sells, and perhaps others triggered by them we see the nett trade position recovers, and by 4pm has reached an excess of some 225K more Buys than Sells. It finishes the day around that level with 2.23M Buys and 1.98M Sells.

Again ascribing only 60% of these, the most suspicious trades to a shorting activity we obtain 90K at precisely 36.0p for the first sequence and 126K at 35.03p for the second sequence.

Thursday, 7th Feb 2019

Thursday the trading had the added interesting feature that the website was down from about 9:15AM to 1:08PM.

It may of course be a complete coincidence that the first suspicious sequence was observed at 9:19AM - a few minutes after the LSE site went down:

Within the space of 71 seconds over 350,000 shares were dropped into the market causing the nett trade position to spike strongly negative for the 3rd day running. By the time that the LSE came back up pretty much all these shares had been bought. Trading for the rest of the day was still pretty sluggish with only a flurry of rollovers (Buys in Green, Sells in Yellow) just before the end of trading lifting the overall volume figure for the day to just over the rather low figure of 1.65M:

Once more the nett trading position for the day finishes positive with 35K more Buys than Sells.

Friday, 8th Feb 2019

On Friday it appears we had slightly different behaviour on opening - not the 50K or 30K Sells, but also £10K sized sells (just under 30,000 shares):

A series of Sells, all less than 10 minutes into trading cause the Ask to be dropped again from 36.0p to 35.0p  Following this the Bid and Ask remained roughly constant, albeit relatively light trading meant that this could be obtained with only a few 50K and one 100K Sell. 

On the chart the period around 12 noon does look superficially like it could be a period of shorting - at 12:04PM the Bid is slashed from 34.50p to 33.0p and there is a strange sequence of trades, placed and then reversed just before noon:

However there is no clear reason why on earth there is such a dramatic impact on the Bid at 12:04 - the reported trades around that time are all very small. We may wonder if there are trades yet to be reported or whether this was just some sort of share fishing exercise. Following strong buying immediately after the nett trade position quickly went positive - however it was again dragged negative by a very strong period of selling between a quarter past two and 3 o'clock:

Once more, however, buying sucks up these Sells, and the nett MM trade position for the day finishes very close to neutral.


Over the week on a total volume of 15.6M there appears to be an excess of more than 600,000 Buys. Despite this the SP fell from 36.75p to 34.5p (6.5%).

This result appears driven by shorting behaviour which I very conservatively estimate at a size of 800,000 shares. Average strike price for these sells is close to 35.5p. 

Following the publication of this analysis if the lse forum sees a sudden infestation of trolls you can probably bet your bottom dollar that they are these actors will be looking to get out as soon as possible on Monday morning. The current buy price for shares is around 34.80p thus it will not take very much buying to shift the SP up past the point at which the shorts are underwater. If they have placed these via spreadbet platforms, which of course avoids any capital gains tax and the need to actually have any shares loaned for the short position, then it may not be much higher that the SP has to go before the more leveraged of these bets receives a margin call. (Basically for leveraged bets the platform chooses when to stop your game for you once you are under water.)

At this point the position will be closed, with the gambler losing out and the spreadbet platform being forced to buy shares in the market. This may trigger further rises in SP, and the process repeats with those less leveraged now being cut off at the knees by the spreadbet platform. The short squeeze gathers pace, like a snowball rolling downhill, eventually becoming an avalanche - almost without any human intervention, since the spreadbet platforms operate purely by algorithm (computer says you lose !)

For LTH's the key thing is to be buying shares right at the start of this process - everyone who comes after will have to pay more, but once the process starts it may not stop until all the short positions are liquidated in this fashion.

This article only conveys the personal opinion of the author. Whilst every effort is made to ensure the content is accurate, we cannot guarantee the accuracy of the data shown. This article does not constitute professional, financial or investment advice and must not be used as a basis for making investment decisions.

Site content is not authorised by the FCA and you are not safeguarded by the Investor Protection measures of the Financial Services and Markets Act 2000. See our full disclaimer