coalman

They don't. It's an average of everyone who has been in that position which can give an indication of a player who may be good at finishing or otherwise. Plus the actual value is decided by a human based on a set of guidelines which is why different sources will give different xG for the same game (as we saw for the Pompey match on Sunday). Even discounting the fact it's an average of averages, the human element means it's about as far from a measure you might use to make real world decisions as you can get. At best it might give an indication whether you were in the game and whether the scoreline is flattering or whether you need to understand why it has happened.

Particularly away - we've conceded more goals away from home than any other team in the Championship.

Armstrong will be out to prove us wrong and put himself in the shop window before it closes. I have a good feeling about this. 2-1 to Stoke with Armstrong scoring a penalty.

Or more numerous. The BBC had xG as 1.66 to 1.11 in their favour on the basis of 14 shots (3 on target) to our 6 shots (4 on target). So while we had the better chances we really didn't create all that much.

Hire a competent manager. Then trust them to resolve the issues with culture, training, tactics, fitness without micromanaging them based on some gambler's spreadsheet from 10 years ago.

I'm a bit sad that Martin didn't get to properly stamp his cultural identity on Rangers. He wasn't there long enough to properly establish the patterns of training, tactics or excuses that were his legacy at Southampton, Swansea or MK Dons. Hopefully Leicester have forgiven him for choosing Scotland over the Midlands and will bring him back to the Championship.

And, therein lies the problem. If you buy enough players then sometimes you're going to find a good one. Is that because of your process or in spite of it? In the same way that recruiting a Sulemana or a BBD doesn't mean your process is bad. Attempting to rank all teams in Europe into a single league to identify the teams doing well is a novel approach but you're comparing vastly differing sample sizes of teams playing between leagues and basing your inference model on performance that changes from week to week and season to season. Are a player's stats good because they are good or is it because of the players around them or the tactics they are employing? Does a recruitment approach that worked for a Danish top tier side work for a Premier League team? This is just scratching the surface on why this approach is sketchy. Fundamentally, is a team outperforming its league down to its players, training or something else? Unless you've demonstrated a causal link you're just making decision based on coincidence (or statistically, correlation). At best, you might use a statistics based approach to identify players that might be worth looking at further. But then you have to watch them (whether on video or in person) to understand why the stats are saying that. You also want to watch them when the camera is not on them (for example - anyone watching Manning play with the ball might think he's a great ball playing full back - anyone watching him amble around the middle of the park without the ball would immediately spot that he's left a gaping hole for the other team to exploit at left back). To give some examples. It's arguable that Pochettino was among our best managers ever. Adkins got us promoted twice and Saints were competing in the Premier League but changing from Adkins to Pochettino led to a marked improvement in both performance and results with the same players. Almost overnight. Hughes managed to keep us up and yet changing to Hassenhutl again led to a massive improvement with the same players. Did the players become that much better or were they in an environment where they were able to perform better? Towards the end of his reign the game had moved on and Hassenhutl's tactics were found out. Does this mean the players had gotten worse? Ankersen's approach has no way to normalise for this. If he'd analysed Saints under Adkins he'd have been far less likely to see Saints players as worthy of recruitment yet under Pochettino the exact same players would have been a far more attractive proposition. One might reasonably argue that getting the manager right is critical to getting the environment right and the best out of players. Under Ankersen's approach the manager is there to implement policy with the players he is given not develop an environment in which they can excel.

Russell Martin has a wealth of experience in the Championship including getting promoted.

Fair points. Maybe that's part of the problem - they squad aren't invested in each other's success.

I think they're referring to this interview with Ankersen where he says he prefers watching match videos over being in person though that the scout's job is not to say whether the play is good or not because they already know. Some key quotes below https://thecorrespondent.com/2607/how-data-not-people-call-the-shots-in-denmark/230219386155-d2948861 Where he says: ‘The scouts do a different job here,’ Ankersen says. ‘We tell them: your job is not to tell us whether a player is good or not. We already know that. The scouts need to see whether they are a fit from a personal, psychological point of view.’ In fact, Midtjylland are questioning how valuable it is for their scouts to travel to live matches to evaluate players. ‘One game doesn’t tell you anything as a scout. If you base your opinion of a player on a few games you’ve attended, it will blur your vision. It’s a small sample. We believe it is more effective to see lots of matches on video.’ Last week, Ankersen travelled to Spain to meet with a player who is on a team in the Segunda Division. According to the model, however, the club plays like a mid-table team in the Primera Division; i.e. a lot stronger than its real-world position. ‘And I’m not going there to watch him play,’ Ankersen says – this could only muddle his opinion of the player, who he knows is good. ‘It’s to convince him to move to Denmark.’ ‘People see huge difference between the Premier League and the lower divisions in England,’ Ankersen explains. ‘We think this is not true. There is a big gap between the Premier League’s number 7 and number 10. But the gap between the Premier League’s number 10 and the Championship, or even League One, is far smaller.’ Sparv landed on the club’s radar because Midtjylland’s model was bullish on his former club, Greuther Fürth, from Germany’s second tier. ‘Our model says that last season, Greuther Fürth were good enough to play in the English Premier League.’ ‘We redesigned the club based on a question: what would a football club look like if it had no human eye and ear? Of course, it turns out you need a human element. But if you say from the start that ‘Oh, it has to be a combination of stats and humans,’ then you won’t be radical enough to be able to make a difference.’

From the replays I saw it looked like he was already berating Leo for shooting before it hit the back of the net. And after the goal went in he just stood there and didn't celebrate at all. No idea why he reacted that way - but it was a really strange reaction.

Armstrong of all people can't complain when someone decides to take a shot. The man lashes at goal every chance he gets. Leo was clean through one on one and I trust his composure way more. Not celebrating a goal against the Skates because you didn't score it isn't a good look.

Positives from that game. Back 4 with Bree and Wellington looks so much more solid. Leo's goal and celebration. That is all. Tonda really failed to get the most out of the better quality we had on the pitch and engaged Pompey in their game of boot it upfield when our best players are all best with ball to feet.

Some kind of tricky winger?

Peretz did really well there

This game is still there for the taking. Where's the "pure desire" Tonda was talking about after Hull?

Trainer watch. Mousinho's previously pristine trainers are looking a bit muddy whereas Tonda is still brilliant white.

Seeing the replay, Armstrong was already berating Scienza for not passing to him.

Good subs. Let's go for this.

It's a completely different feeling than when Armstrong is clean through.

Leo!

A good thing about both Bree and Wellington is they're going to put the player trying to cross the ball under pressure.

Ref needs to find his cards.

The difference is that's their game plan. We should have more about us.

I'd bring Fellows on straight away. Matsuki is leaving big spaces down our right hand side. I might move Matsuki inside and replace Azaz though. And would definitely bring Charles on for Downes. Stewart for Armstrong needs to happen at some point if Stewart is up to it. Not unhappy with our defence which is something I haven't had a chance to say much this season.