The first point I made about creative defenders is essentially what would later become known as “flying fullbacks”. 

This phrase was popularised in October/early November after Alexander-Arnold’s incredible burst of form that month. That’s a full 3 months after my July analysis.

In October, you saw lots of “analysis” starting to appear on social media suggesting a huge change in strategy or a renewed focus on this “revelation” of flying fullbacks as if it was something new. “If TAA can do this, who else can!?”

But that is a very bad time for analysis.

The difference between what I did and that is timing – I can work out what is very likely to happen by using the new matrix, match data, and my own judgement from thousands of hours of experience. 

By contrast, if you are waiting for the FI performance scores to highlight who is looking strong – it’s too late. 

If someone thinks they are doing well by picking up flying fullbacks in October, imagine how well the guy who had a 3 month headstart will have done.

And this form of analysis has a second trap that keeps traders into an overpriced trade longer than necessary or leads them in to paying too much for a decent player – anyone relying on historic scores and dividend yields is also by extension assuming that TAA’s outstanding October performance is normal. 

To this day, you still see justifications of his price based on his historic “dividend yield” that they attempt to project forward over his career. 

This is absolute garbage and an overreliance on these historic metrics will lose us money and possibly some brain cells.

In October (and early November actually) TAA’s numbers were through the roof not just for himself but he was streets ahead of any full back in the big 5 leagues. 

It was so good, it felt a little too good to be true as covered in Scouting at the time. Things that look too good to be true generally are (See Haaland’s recent goals tally). And it has dipped on and off throughout the season as we might expect. Maintaining that level or even close to it was a near impossible ask.

And, there is a huge slice of luck in dividend wins anyway. Your boy can score 250 and lose to a random Metz Centre Back who has the best game of his life. He can score 190 and get extremely lucky and win a dividend. And then the randomness of game winning goal too obviously.

This alone makes dividend yields and “overs” (number of scores over X) unreliable. Especially since a good “over” still seems to be considered 200+ in many circles. This is a mental hangover from the old scoring system – a 180-200 doesn’t mean much at all anymore. 

We at least want to be into the 220’s now. Traders are, on the whole, too easily impressed.

Then, factor in all the footballing reasons why our player’s output might change over a career and even 5-10 games from now. New tactical instructions. New coach. New team. Maybe other coaches figure out TAA is a weapon too dangerous to be ignored and find a way to shut him down. 

The main point here is that recent match data is the best primary source when conducting analysis. 

Historic performance scores are very much secondary. I look at them to see whether what I have learned from match data is appearing in the actual scores. This is interesting but is very much influenced by luck. I never look at historic scores as a way of judging ability.  

Things like dividend yields? I do find them of interest but only because a certain type of trader will over use them and I can predict what people will buy when they use this data. 

I may for example buy an off trend player with a high dividend yield whilst they are cheap in off-season, knowing that when dividend returners eventually come in fashion again when football is back, people will tend to buy the high returner and become less sensitive to whatever reason they are off trend (i.e age). 

But because I have a better idea of the players current real ability using match data I have an advantage. I can choose to sell at the higher price if I know they have won more dividends than they really have a right to, for example.

Bear in mind too that I am actually a TAA fan believe it or not.

Go to my July 2019 analysis of the most likely to benefit from the scoring changes and TAA is there. Or my player ratings tables where TAA is the only defender on there.

But you can pay too much for good players.

Look at TAA’s price chart below. If your analysis is on point in July from the new scoring system, you’ve seen evidence that creative fullbacks are going to do well and you’ve pinpointed TAA as one of them.

If you want to make that bet, July to mid October at a push is your moment. When it’s something you know is very likely, but it hasn’t been proven yet.

Once he starts winning regularly in October and November and social media has blown up with it’s “Flying Fullback” phase – that’s too late to buy. That is when I would suggest selling. Why? Because we know he is good but we know he is likely not £6+ good.

A good analyst got a 3 month advantage here and didn’t have to wait for matches to tell them creative fullbacks would do well. 

We could not predict the October hotstreak in an individual at all, clearly. But we could buy up a selection of 5-6 promising full backs based on our July research and then let them run. 

Chances are, one of them would probably put a good run together. And then the rest may get pulled up too once the search for “the next…” starts.

Eventually, TAA’s performance will have to increase or his price will have to come down. Either of those is possible but given what we know from match data, we know he isn’t likely to get much better because he’s been outperforming every full back on the continent as it is, and we know he’s had a few lucky wins with relatively low scores. 

It is far more likely therefore that the market is overvaluing him than not. 

Of course, it is possible that he goes on another hot streak and that tantalising possibility is what keeps people in. More likely though, he’ll keep performing well but not quite justify the huge price tag and eventually slide down to a more sensible value.

In my view, we want to play the percentages here and put ourselves on the right side of probability. Especially considering that we should have much juicier targets in the £1-£3 range who are much more likely to increase by a greater percentage value from here.