The9

Only one in 5 Corners even lead to a shot on goal, and only 1/19 free-kicks leads directly to a goal - they're specialist areas and it's good to improve them, but their impact is significantly less than most people think.

Whoscored had a whole ream of stats showing Pelle was doing well when people were slating him last season. I put up stats showing he was better than Diego Costa in aerial duels and people just decided to ignore it. Was it Drogba who came top?

I'm still fairly sure it was losing key players all over the team from the Chelsea match onwards, and not having Villa being comical against us every week. Confidence is a fairly intangible concept, interceptions, blocks and tackles aren't: http://www.premierleague.com/en-gb/news/news/2014-15/nov/131114-defence-key-as-southampton-continue-to-shine.html In November 2014 Saints had made 0 errors leading to goals and 0 errors leading to shots, had only allowed 24 shots on target in the entire season (2.2 per game, leading the Prem). Only 91 shots faced, 2nd in the league, and Forster only had to make 18 saves, 2nd lowest in the division at the time. We had also only used 6 players in the back 5 in that time. The midfield had made 10% more tackles than the next-best team in the league. All of these stats show that we allowed less efforts on goal, had a stable team and prevented chances being created. Even the stat showing Forster had the best save percentage at the time (77.3%) doesn't prove anything much when you consider how few saves he actually had to make, and how the other stats show that teams weren't getting good quality efforts on target. Now consider that we removed Schneiderlin (himself Saints most effective player in terms of goals scored vs conceded when on the pitch), and/or Wanyama, and moved Alderweireld from the back 5, replacing him with Yoshida (who comes last in the net scored v conceded list last season with 0). How is that NOT going to have an impact? It already did have a huge impact against Stoke, Sunderland, Leicester et al near the end of the season, Saints need to be sure that the DM/CB gaps are well plugged before the season starts.

How do you know he hasn't got a levitating lift like in Charlie and the Chocolate Factory?

I can't imagine Saints are going to hazard an untried CB who isn't clearly the greatest thing we've ever seen already in the youth team. There's just far too much risk involved. This is just the Guardian looking for a position where we're short and assuming we'll plug some kid in as a stop-gap because we're "only Southampton" and have a history of good academy projects. Plus they can't suggest Stephens or Turnbull because people actually saw them getting dissected in the L1 Play-Off Final. Also the pedant in me is screaming "Premiership? Is he going to Scotland?".

I demand an ITK snippet sticky for this!

The very nature of the thing means they don't want fans there, that would be a very good way of ensuring they would have to deal with that problem.

Ok, "seems to indicate a high probability of correlation". I'm just getting into this stuff, trying not to alienate the wider audience.

I call that the Rodgers Effect.

I think a lot of this black box talk underestimates just how much of this kind of stuff clubs capture. The key of course, is making sense of it so that it actually gives you useful intel and not just a pile of confusing numbers, or data that shows no causality.

Cheers, I try to help, when I'm not being annoyed by everything. I suspect given my recent delve into analytics that I'm going to get preeeetty tedious in some other areas though...

Schneiderlin was an attacking midfielder when we signed him, the only reason he got to play the way he wanted was because we'd trained him to be that kind of player. Van Gaal is signing a near finished product and the only thing he'll need to do differently is adapt to a poorer defence behind him, which he can pretty much only do by being as good as he already is. His stats may be less impressive next season as he's probably going to have less to do with United being more likely to have the ball in attack than Saints.

Yeah, I'll be there, and Eastleigh will bloody slaughter them. I've actually got £50 on Newport to go down this season, and a free bet to the same amount added to it. It'll be a crap way to pay for a trip to see Saints in Europe come May. I like how that (technically) gets a place on the European match thread.

In case anyone wants a Salzburg programme (in German, 32 pages though) there's one here: http://pages.ebay.com/link/?nav=item.view&id=261967487281&alt=web

One of the stats picked up by Anderson/Sally was that Wigan, who were given as an example of a team bucking the trend of wages to position for 7 years in the Prem, used to shoot more from outside the area than any other club. They targeted players with long range shooting abilities and chose a strategy with a low percentage of success but a high return when it was successful - that allied to their success on the counter and at free kicks helped them stay up against the odds for years. Bear in mind the odds of them getting relegated most seasons based on wages alone was around 60%, and they beat that for 7 years. The reason Saints don't shoot from outside the box is that for every additional successful pass the likelihood of a goal being scored becomes greater. Also, reducing turnovers in possession is a very effective way of preventing the opposition from scoring.

Because we pick U21s to bolster the bench, and they don't need to be counted in the figures.

Looked like fair comment to me. Stephens is likely to be behind 4 others so is hardly likely to have a huge positive impact on results from there.

Small sample size but very useful indicator of how effective we were. Probably a figure people should bear in mind every time they complain that we didn't score enough when that balance is the obvious trade off.

But why are you looking at the goalkeeper and not the weaker players in the team who weren't as effective as their predecessors in preventing the opportunities? It's all about identifying preventative actions that mean the keeper doesn't have to make the save in the first place.

Depends on your definition of "weak", but with us seeded in 3rd Qualifying Round and our opponents not, we'd be unlucky to face a better side in that round - Hull rested players in EL last year and lost in the Playoff Round, and we'd have to be lucky with the Playoff Round draw not to face a difficult opponent with 15 Champions League losers dropping in as we probably won't be seeded for that one. So maybe 2 (3rd Qualifying Round) games to test a couple of them out, then the Playoff Round. We may also have some dead rubbers or possible easier opponents in the Group stage if we get that far, but let's cross that bridge when we get to it...

Actually I think the loss of Forster was hugely overstated. Can't prove it without figures, but Forster's injury coincided with the loss of Schneiderlin and/or Wanyama and as a consequence Alderweireld played in midfield where he was less effective - a quick check of the stats for opposition shots allowed in those games compared to the season average would probably give a decent indicator (as would a stat for defensive interceptions). Keeper shot/save percentages might show the overall effectiveness of the different GKs but at the same time, that doesn't give a measure of the quality or proximity of shots or likelihood they'd be goals whoever the keeper is. We'd need a baseline GK save figure for the whole League over a long period - and Davis/Gazzaniga's sample size isn't exactly huge, so the enormous luck factor involved in all football could be the overriding factor.

Previous transfer activity in this area points to this not being the case. The Alderweireld Experiment was a qualified success, only qualified due to his injuries making him less available. Whenever he played CB you could see how much better we were defensively - the stuff about anticipation and positioning was obvious even without needing the stats. We have also repeatedly signed CBs and not tried to bring them through - Forren, Yoshida, Lovren, Alderweireld, Gardos have all been signed in the 3 years we've been back in the Prem. That's nearly 2 CBs a season to add to Fonte and, originally, Hooiveld's presence. Around half of those named players have been Prem successes for us, but we've not once tried to pick an U21 centre back in the Premier League.

Yup. Hope we don't get the Kazakhstan/Armenia tie winners or might need Wednesday as well...

"Depending where it is", is basically your measure. It'll depend on how far it is, who we're playing, what transport is like and the size of their ground and our allocation. I think you may have to be quick if you have no aways from last season but there will probably be some going to general due to the need to confirm travel etc. and people faffing as they arrange with their mates.