Expectations (hopes) with 1/3 of the season gone...

[Hemingway]

Interesting, but you can prove Simon Mignolet is a very good keeper with stats.

I use them to prove the opposite.

Actually, even though I know that was tongue-in-cheek, it's worth saying that most goalkeeper stats are notoriously pointless, so people should generally not use them to try to prove things, and the very few goalkeeper stats that we think aren't pointless actually make him out to be pretty mediocre. Anyway.

Does this look at how the other teams compare from the point Klopp took over or are they from the start of the season ?

Sorry, should've included that. Everyone else is from the start of the season. There was no real reason to make that break for anyone else other than us. There's good reason to believe that our last six games are much more predictive than our first eight, but that doesn't apply to the other teams (except the others that changed managers, but who cares about them?).

[redmike65]

(can we have festive icons all year - they're fun! - )

[Montana]

We're gonna win the league and the capital one cup

[baitman]

the key to us going that bit further and making 4th or better will be our home form and the home fans coming alive again.
i think the away cup game v stoke will be a great game to win and could then be a catalyst that will heighten expectations for the return leg at Anfield.
if the crowd are then up for it and we end up winning then this might give the team and home support the momentum we need to make Anfield rock again.

[RedReet]

Neil Atkinson from the Anfield Wrap wrote in the Echo the other day that his expectations had no ceiling, but also no floor. Probably a good way to look at the rest of the season.

Ha, I'm pretty sure it was him that made the point about winning the league that I referenced.

Also: wasn't sure where to put this, but this seemed the proper place because we're talking about expectations. For those who might care, a quick tour of where a few models say we are and could go, using data from the start of the season and from Klopp's games only (just in the PL; I debated adding in EL and cup games to get the sample size up, but the data's not as available and I didn't want to do it if I couldn't do it for every team and every metric. Just know that if all of Klopp's games were included our numbers would tick a little higher). If you hate stats in football, just skip past. There's a graph and explanation of the stat under the spoiler.

Team Rating
Total: 5th
Klopp: 3rd

This was a metric created by James Grayson, a combination of the next two metrics (TSR and SoTR) plus a third, called PDO, that measures save% and conversion%. It's roughly as predictive as expected goals models in the PL, and much easier to calculate. I went through these ones myself, with the shots data coming from WhoScored.

TSR
Total: 3rd
Klopp: 3rd

Pretty simple: shots for / (shots for + shots against). It's the proportion of shots in a game that a team takes. "Shot dominance" is another term used for it and is probably more illustrative.

SoTR
Total: 3rd
Klopp: 3rd

Same as TSR, but for shots on target only.

Caley xG (shots)
Total: 4th
Klopp: 4th

Each shot is assigned a probability of becoming a goal based on historical data of shots from its location and a few other factors (headed/non-headed, assist type, etc.). Note that if you included Klopp's EL games, we'd improve but not enough to make up the ground on Spurs.

Will's xG (state of play)
Total: 3rd
Klopp: 3rd

A state-of-play xG model: instead of each shot being assigned a probability of a goal, goal probabilities are derived from passes and ball position. I prefer a this sort of model for its nuance: the Ibe chance against Swansea, for example, had a fairly high probability of becoming a goal but never resulted in a shot, so it went unrecorded in a shots-based model. In a state-of-play model, the chance is accurately reflected. That said: turning possession into shots is a skill, so this sort of model does miss something. But early on when we don't know if a team has a fundamental strength/weakness in possession-to-shot conversion, it's probably more helpful. He says we're better under Klopp, but not by enough to catch City/Arsenal.

So, for those who just looked at that and don't really know what to take away from it: Klopp has very definitely made us better, we're probably one of the best three or four teams in the league, we're right in the thick of it for top four, and the title is probably a little beyond us right now.

To win the title, we'd have to not just play to the level Klopp's had us at so far, but also keep getting better, and hope that City's first eight games were a fluke and that Arsenal fall apart. The last is likely given the injuries; I'm wary about City, though. I just don't see any reason to believe their last six games, during which they've been pretty average, are any more predictive than their first eight, when they were the best team in the league by a mile. Even if they sustain a period of poor performance, they'll just sack Pellegrini and bring in an older head on an interim basis while they wait for Pep, and changing managers midseason usually has a reset-button effect that would be enough for them. They don't need someone to come in with a revolution, like Klopp; they'd just need someone to get them playing to their baseline and that'd probably be enough for the title. Catching them will be tough, and require them playing along.

And then there's United. They're not in the top three in any of these. In SoTR, they're below Palace. They're sixth in Caley's xG model, seventh in Will's, even below Chelsea. We don't know that much about analytics in football, but what we do know tells us that United are mediocre and should fall back to earth any day now. But there they are, winning football matches, sitting in third, a point off the top. To challenge for fourth, let alone the title, we'd need them to regress to where the numbers say they should be at some point. And frankly I have no idea if they will, or if they do when that'll be. They play a style of football hitherto unseen, at least not a style we have any data on. It could well be that Van Gaal has gamed the system, and he's doing stuff that the models aren't picking up on. But then that's what everyone said about Lucien Favre, too, and look what happened there. Whether it's luck or just very successfully doing very hard things and walking a tightrope, I'm going to assume that at some point they start to fail. I just don't know how long it'll take.

Anyway: after all this, I'm sticking by my original assessment, which is that I'm happy enough with fourth and a cup run. Maybe we do keep getting better, maybe City is complicit in their own downfall, and maybe United does finally fall away like they should. Won't rule it out, won't be disappointed if it doesn't happen. No ceiling, no floor.

Interesting. Small sample size like you say and two ways to look at it. Are we over-perfoming due to the new manager effect or is there more to come as the players get more used to Klopp? In saying that, I think we were six points behind when Klopp took over so maybe it's neither.

[Buzzo]

Neil Atkinson from the Anfield Wrap wrote in the Echo the other day that his expectations had no ceiling, but also no floor. Probably a good way to look at the rest of the season.

Also: wasn't sure where to put this, but this seemed the proper place because we're talking about expectations. For those who might care, a quick tour of where a few models say we are and could go, using data from the start of the season and from Klopp's games only (just in the PL; I debated adding in EL and cup games to get the sample size up, but the data's not as available and I didn't want to do it if I couldn't do it for every team and every metric. Just know that if all of Klopp's games were included our numbers would tick a little higher). If you hate stats in football, just skip past. There's a graph and explanation of the stat under the spoiler.

Team Rating
Total: 5th
Klopp: 3rd

This was a metric created by James Grayson, a combination of the next two metrics (TSR and SoTR) plus a third, called PDO, that measures save% and conversion%. It's roughly as predictive as expected goals models in the PL, and much easier to calculate. I went through these ones myself, with the shots data coming from WhoScored.

TSR
Total: 3rd
Klopp: 3rd

Pretty simple: shots for / (shots for + shots against). It's the proportion of shots in a game that a team takes. "Shot dominance" is another term used for it and is probably more illustrative.

SoTR
Total: 3rd
Klopp: 3rd

Same as TSR, but for shots on target only.

Caley xG (shots)
Total: 4th
Klopp: 4th

Each shot is assigned a probability of becoming a goal based on historical data of shots from its location and a few other factors (headed/non-headed, assist type, etc.). Note that if you included Klopp's EL games, we'd improve but not enough to make up the ground on Spurs.

Will's xG (state of play)
Total: 3rd
Klopp: 3rd

A state-of-play xG model: instead of each shot being assigned a probability of a goal, goal probabilities are derived from passes and ball position. I prefer a this sort of model for its nuance: the Ibe chance against Swansea, for example, had a fairly high probability of becoming a goal but never resulted in a shot, so it went unrecorded in a shots-based model. In a state-of-play model, the chance is accurately reflected. That said: turning possession into shots is a skill, so this sort of model does miss something. But early on when we don't know if a team has a fundamental strength/weakness in possession-to-shot conversion, it's probably more helpful. He says we're better under Klopp, but not by enough to catch City/Arsenal.

So, for those who just looked at that and don't really know what to take away from it: Klopp has very definitely made us better, we're probably one of the best three or four teams in the league, we're right in the thick of it for top four, and the title is probably a little beyond us right now.

To win the title, we'd have to not just play to the level Klopp's had us at so far, but also keep getting better, and hope that City's first eight games were a fluke and that Arsenal fall apart. The last is likely given the injuries; I'm wary about City, though. I just don't see any reason to believe their last six games, during which they've been pretty average, are any more predictive than their first eight, when they were the best team in the league by a mile. Even if they sustain a period of poor performance, they'll just sack Pellegrini and bring in an older head on an interim basis while they wait for Pep, and changing managers midseason usually has a reset-button effect that would be enough for them. They don't need someone to come in with a revolution, like Klopp; they'd just need someone to get them playing to their baseline and that'd probably be enough for the title. Catching them will be tough, and require them playing along.

And then there's United. They're not in the top three in any of these. In SoTR, they're below Palace. They're sixth in Caley's xG model, seventh in Will's, even below Chelsea. We don't know that much about analytics in football, but what we do know tells us that United are mediocre and should fall back to earth any day now. But there they are, winning football matches, sitting in third, a point off the top. To challenge for fourth, let alone the title, we'd need them to regress to where the numbers say they should be at some point. And frankly I have no idea if they will, or if they do when that'll be. They play a style of football hitherto unseen, at least not a style we have any data on. It could well be that Van Gaal has gamed the system, and he's doing stuff that the models aren't picking up on. But then that's what everyone said about Lucien Favre, too, and look what happened there. Whether it's luck or just very successfully doing very hard things and walking a tightrope, I'm going to assume that at some point they start to fail. I just don't know how long it'll take.

Anyway: after all this, I'm sticking by my original assessment, which is that I'm happy enough with fourth and a cup run. Maybe we do keep getting better, maybe City is complicit in their own downfall, and maybe United does finally fall away like they should. Won't rule it out, won't be disappointed if it doesn't happen. No ceiling, no floor.

Seems a particularly relevant post this now Ernest 'No ceiling, no floor' I couldn't agree more.

[captainfog]

Top 4. With a bit of luck top 3.

Same.

[G]

Comfortable top 4 , win a cup, within a few points of top spot. We've improved so much in a short space of time and now we are getting everyone fit I think we've a great chance of being in the mix. Just like 13/14 I'd much rather enjoy the ride and believe we are in the hunt than be all Sarb-ish and refuse to let myself believe.

We have Klopp, the quadruple is on

Same.

Nobody is owning the division and nobody is clearly miles better than us.

Top 4 is well within reach and if we sort out the home form and keep Sturridge fit so much more is in reach too. Even after today, only 3 teams gained on us and one of those defo won't last the pace.

[Phoenix07]

A challenge for 4th, and a decent showing in the cups (depending on the draws).

As shown today against Newcastle, the moment we're hyped up, teams will raise their performance levels, and it's up to us to show we have the answers. Today, we didn't. I feel that the squad is beginning to understand the demands of playing so many games, but they're still in their infancy in terms of this development. Having to come up with the answers to different questions every few days can be tiring physically and mentally. We'll be better for it in the long run, and I'm ecstatic we have Klopp at the helm to manage us through this.