18th Aug Flash Short – Week 9

Alfacomp

22 October 2017

Monday 16th Oct 2017

An apparently quiet day with no movement of the share price, despite being punctuated by  SOME BIG TRADES.

Can it be a PI dropping 250,000 shares at 8:15 AM, and then again at 11:29AM and then 500,00 shares at 11:55AM ? No. Can it be a shorter ? No. How can any of these people get a price of 9.50p for any sort of trade of 500,000 at 11:55AM ? They cannot – this must surely be one market maker selling to another to allow the recipient to close a short position on their books.

Whilst the apparent nett trade position reaches -1.15M at the end of trading this incorporates the 1M in trades described above which I do not believe represent true PI Sells to MMs but instead MM1 Sells to MM2. Thus the nett trade position, in these circumstances, overestimates PI Sells. I think it is clear on this particular day, but more generally it is harder to distinguish the inter-MM trades that they use for balancing their books from the real MM<->PI trades that they can masquerade as.

The SP ended at 9.625p, flat on a volume of 1.40M (122K Buys vs 1.274M Sells).

Tuesday 17th Oct 2017

A genuinely quiet day without share price movement or any obvious inter-MM balancing going on. The nett trade position went positive early on after a 200K Buy (from one of the well-known LTHs) and remained there for the rest of the day, until a few minutes before closing time when a couple of Sells popped up to take this back down to zero.

The SP ended at 9.625p, flat on the day on a volume of 663K (306K Buys vs 357K Sells).

Wednesday 18th Oct 2017

A simple glance at the plot above reveals that something seemed to happen around 12:45PM – but what was it ? The first sign was the strange sign that the Ask dropped from 10.0p to 9.95p – this was as a result of one particular Market Market (CFEL – Cantor Fitzgerald Europe) dropping their Ask below the 10.0p of all the other MMs. This was then followed by a different MM dropping the Ask to 9.75p (CFEL retained their 9.95p Ask until friday) – this coincided with a Sell right at the bottom of the Bid-Ask range at 9.25p – was this a big Sell by a PI ? No – it was only 25K, so hardly earth shattering. There is no obvious explanation for this small move in the Ask.

Trading picked up in the afternoon with buys after 2PM exceeding sells by some 350,000. It is interesting to see how some of these buys fall at 9.6p and some at 9.7p, but nowhere in between – possibly a result of having two different market makers offering Buys (one with Ask 9.75p and perhaps the other, CFEL, with Ask at  9.95p) – whoever was getting 9.7p for 61,726 shares at 3:01PM seems to be getting a bit of a raw deal.

The Vanitec 3rd ESC meeting – Part 1 article was published at 5:30pm – which strangely prefigured the arrival of our regular Troll, whose delightful commentary is, alas, no longer present in the LSE record.

The SP ended at 9.5p, down 1.2% on the day on a volume of 860K (534K Buys vs 326K Sells).

Thursday 19th Oct 2017

Thursday started with a couple of small Sells and then a 300K Sell and 100K Sell, which by 9:20AM had knocked the nett trade position to more than minus 500K. Trading continued with mainly sells until a large Buy of 500K (not reported until the end of the day naturally) at 11:32AM. This brought the nett trading position back close to neutrality. The next Buy, at 1:16PM, was quite a lot higher than the 500K trade at 11:32AM, and it was for less shares – thus we might suggest that the first trade (the 500K at 9.50p) was perhaps an inter-MM balancing trade. Usually we ascribe these to sells, but this time as we are drawing the A-D line slightly below the 9.5p Bid-Ask midpoint, it gets put down in the Buy column – we should really start thinking about having a third column  – MM Balancing, to separate these from genuine PI Sells or Buys. Still, putting this down as a Buy perhaps offsets many of the Inter-MM trades that have been previously put under the PI Sells heading.

The A-D line has been lifted in the afternoon – why ? As you can see from the plot the trades fall into three bands – the high band (PI Buys), the low band (PI Sells) and quite clearly on this day – the middle band – close to the Bid-Ask midpoint (9.45p – 9.55p) which I believe are inter-MM balancing trades. The present approach (which we’ll call the simple A-D method) only divides trades into Buys (by PIs) and Sells (by PIs) with the distinction being made by the A-D line – those above are deemed Buys, those below are deemed Sells. This approach has no way of dealing with the Inter-MM trades which otherwise masquerade as PI Sells/Buys. In future we’ll have to think about a more sophisticated analysis but for now we’ll try and not knock the nett trade position (PI Buys-Sells) too much out of whack by putting half of the clearly inter-MM trades above the A-D line and half below.

There was quite a lot of confusion on the LSE board about whether shares bought today would qualify for free Afritin shares (Answer Yes – see ‘FAQs – Will I get free Afritin Shares‘) and whether it was too late to vote (Answer Yes – the proxy deadline was the previous day). There were also the remaining parts of the Vanitec 3rd ESC meeting – Part 2 and Part 3 reports published.

The SP ended at 9.5p, flat on the day on a volume of 2.43M (1.16M Buys vs 1.273M Sells).

Friday 20th Oct 2017

Afritin Demerge Vote day – unsurprisingly there was some significant volume buying ahead of the vote announcement (12.55pm) – the vast majority of these before 10am – conspiracy theorists take note.

The Nett Trade position went positive following a Buy of 200K at 8:28AM and remained in positive territory for the rest of the day, increasing to almost 1M nett excess Buys at 3:44PM. The Ask very briefly returned to Cantor’s 9.95p in the morning, then ticked up to 10p soon after the positive Afritin Vote announcement.  Lack of any Sells after 2pm required the Bid to increase to 9.75p at around 2:50pm – this yielded the smallest spread (0.25p) that we have had in the last couple of weeks.

The SP ended at 9.875p, up 3.7% on the day on a volume of 2.66M (1.726M Buys vs 936K Sells).

Summary for Week 9

SP Change TBP estimated Buys – Sells Potential Shorting Activity Comments
Monday 16-10-17 Flat – 1,152,346 c. 1M Inter-MM Trades ?
Tuesday 17-10-17 Flat – 50,990
Wednesday 18-10-17 – 1.2% + 207,379
Thursday 19-10-17 Flat – 115,997 c. 800K Inter-MM Trades ?
Friday 20-10-17 + 3.7% + 789,973
Total + 2.5% -0.32M

Conclusion

 Week after 18th Aug Short TBP estimated Buys – Sells Weekly SP Change
Week 1 – 0.36 M
Week 2 + 1.2 M + 7.9%
Week 3 + 3.6 M -6.1%
Week 4 + 0.76 M – 2.5%
Week 5 – 1.36 M – 1.2%
Week 6  – 1.90 M + 1.2%
Week 7 + 0.09M + 1.2% 
Week 8 – 1.30 M + 1.2%
Week 9 – 0.32 M + 2.5%

For the fourth week in succession the SP has risen whilst at the same time it appears that there have been more Sells than Buys – clearly PIs cannot really be selling if the SP keeps rising so what is happening ?

It seems likely that this is the closing of a short position held by one of the Market Makers – with shares being sold to them by another Market Maker. The short can is just being kicked down the road – raising the obvious question ‘how long can that go on ?‘ This is a subject that we shall no doubt return to in future, perhaps with the benefit of a more detailed analysis of Inter-MM trading, but the general question is can a short be simply booted back and forwards between different MMs, each saying “I’ll deliver you your shares in three (or X) weeks” – surely the PI who bought the shares in the first place has to get their shares ?

You might think so – in the real world of when you buy something (say some apples) you expect to get that thing (so you can have lunch), and not just a promise that you’ll get something from Mick, who’ll get it from Ted, who say’s he’ll get it from Tom ….

But think about how a bank works. You pay in real 5 pound notes, with serial numbers on them – the bank takes them and then tells you you have credit. Go to a bank and ask for your money – you usually get it, but on occasion you do not because the bank needs to get them from another bank – who needs to get them from another bank and so on – you see the similarity. Whilst in principle this can go on forever it does not because banks have rules about how much real cash (and assets) they have to hold to back up the promises that they give you when you hand over your notes.

The same will go for Market Makers – they will need to ultimately limit any short position that they can temporarily have on the books, both in size and time. Eventually those other MMs that they need to buy their balancing shares from will up the price that they are prepared to offer their buddies – eventually the short will work out, but only once the SP has risen sufficiently for one of the MMs to buy the shares that they need from real PIs.

The simple answer to the question “how long can it go on ?” is therefore “not forever, but long enough to screw sufficient shares out of those who have the least patience.”

The Graph above shows the SP and the daily nett trade position shown in the 9 weeks since the 18th August flash short. This is calculated using the simple A-D method of distinguishing PI Buys and Sells. Further analysis is underway to separate Inter-MM trades from these to give a more revealing picture of the market dynamics post this flash short attempt.

 

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