A poster on EST1892 called Hemingway made this post which was very interesting I thought. Just to give context, this post was made as a part of the discussion on Brendan's statement that he was now a better coach than 2 years ago under a title challenge.
"Actually, Brendan probably is a better manager, in some unquantifiable holistic sense. People tend to improve with time and experience, so he's had to have gotten better in at least some respects. The problem is that managers aren't judged in that unquantifiable holistic sense, in which space exist "good managers" and "bad managers." There are a few genuinely good managers and a few genuinely bad ones, but a whole lot in between who have only a fleeting impact on their club's success and are dependent on how they fit the club's situation (and getting some good fortune). Brendan exists in that group of not-bad, not-great managers - he had a good run in a good situation, and has since found out that the particular plan he used in that season just doesn't always apply.
I've made this point before (as has Lecter, though more critically of BR), that the title challenge year couldn't be used as a blueprint because it was dependent on an unusual number of circumstances that weren't sustainable. Just off the top of my head, it required:
- having two world-class strikers;
- having those two world-class strikers outperform even their very high standards;
- scoring an unusually high number of set piece goals, many of them coming early to break open games;
- outperforming things like [statto alert] TSR, SoTR, and xG, which usually come back to earth with time;
- hell, even getting an historic number of blocks (which, like set piece goals, regress to the mean season-on-season).
I'd argue that all five of those things were unsustainable. Statistically, we know the final four are. You can beat them for awhile, but you always regress to the mean given enough time. But I also think that having two world-class strikers is likewise unsustainable, and people gloss over this. How many clubs can consistently have two world-class strikers? Real, Barca, PSG, Bayern...maybe United? Unless you're one of those really moneyed clubs, any plan that relies upon having two strikers of Suarez-Sturridge levels is pretty much bankrupt. If we held onto Suarez, or if we replaced him better, we could've delayed the downfall for awhile. But it would've come eventually. Except for that really elite few, it's an exceptionally difficult thing to maintain.
A lot of people say the downfall from that season started when we tried to find another plan, instead of just getting goals on the pitch and seeing what happens. I disagree - the first aspect of that plan would've been hard to maintain and the other four were just straight-up statistical quirks that would've (and did) come back to earth. It was unsustainable to do all of that year on year. We were right to look for another plan; the problem was that BR couldn't make another plan work.
That doesn't mean he doesn't deserve credit for that year, or that he can't find another method that might work in the future. I'm hopeful about 3 at the back, actually. But you (and he) can't just take what he did in 13-14 and say "See, he can do it, and he's even better now." No, he can do it if he has two world-class strikers and things break right. Which is not the same thing as being able to do it in the current situation. He's better, but not as lucky. He's better, but in a worse situation. And right now it's looking like he doesn't have the answers for it.
Anyway, he's not a bad manager, certainly better than the last month or so, and I wouldn't be surprised if he goes on to be successful elsewhere once he gets some time away from the game to reassess things. I'd be happy enough if he could figure it out here. But I don't think he will, and based on his history of making grandiose statements before important games I'm not optimistic about getting the results he's hoping for."
This was the follow up post
"Sorry, should've explained in the post. TSR is total shot ratio (shots for / shots for + shots against), basically the ratio of shots taken in a game that are being taken at the right end. SoTR is the same thing, but just for shots on target. And xG is expected goals, which takes shots, looks at their locations and how they were taken (the type of pass that led to them, the number of passes that led to them, and so on) and returns how often you'd expect a goal to be scored from that shot, based on historical data. They're all based on the idea that finishing tends toward randomness - it's easy to have a hot or cold streak that covers up how teams are actually performing - but that over the long-term teams will order based on how many shots they're getting vs. allowing, and the quality of those shots. None of the three are flawless, but they're pretty good, consistent indicators of performance, and over a long period of time (a couple of seasons, for instance) you can expect almost all clubs to perform to them, with a very few exceptions of teams who are doing stuff the numbers don't catch. They aren't fancy, but they're the gold standard as far as public domain stuff goes and are fairly predictive.
We were tied for fourth in TSR and were third in SoTR in the title challenge year. James Grayson has a calculation of those two combined with a third metric, PDO (which is more confusing) that put us third in that season, closer to 5th than 1st. We were second or third in expected goals difference depending on whose model you look at, but overperformed expected goals scored substantially. (There was about
an 8% chance of scoring as many goals as we did from the shots we took - the two world-class finishers account for some of that. We actually underperformed expected goals against, which evened things out a bit.)
Effectively, all this just means that we should've expected to fall some anyway, even if
nothing changed - which I think was totally unrealistic - because that year we were doing things and getting results that clubs struggle to repeat season-on-season. The set piece thing, for instance, was never going to happen again, and that was a huge part of our success that year. The season wasn't a fluke, but we did do better than we could be expected to do in the future, as
many said when it was happening. I was a pretty big sceptic at the time but that's exactly what happened and I think it helps explain some of the decline experienced since."