Monday, 19 May 2014

Wenger and Wonga

Now is a time for celebration, even some modest crowing. I predicted we would finish the season on 78-80 points, and we ended up on 79. Whoo-hoo. I also thought that Man City were the most likely of the top teams to pull away and clinch the title, though I hadn't expected that they would leave it so late, and that Aaron Ramsey would make the key contribution to Arsenal's fortunes at the death, though I confess I was hoping more for a cheeky "Michael Thomas" than an "Alan Sunderland". My other prediction, that Wenger would commit to a new contract, now looks odds-on, but I always felt this likely regardless of silverware or league position. He will have given his word in advance but insisted on seeing out the current contract. That gives the club some wiggle room, but they're clearly not scouting for a replacement.

The FA Cup final was short on quality, as it often is, but the two exceptions to this, Cazorla and Ramsey's goals, made us fitting winners. Once the Spaniard had opened our account, I was confident that we would win, notwithstanding an error-prone referee. For all the drama and Hull's fight, this was a technically accomplished display of how to wear down an obdurate opponent, something we have had to do many times in the league. Wenger is routinely accused of being tactically inflexible and incapable of making impact substitutions, but his interventions were spot on. The perennial focus on character and belief also proved sound. Mikel Arteta even laid to rest the myth that Wenger doesn't talk at half-time (his critics usually advocate hurling crockery and shouting "Come on!" till veins pop).

Despite the "new chapter" effect of silverware, I suspect the media template for next season was fixed after the defeat at Anfield, much as that also rewrote the Liverpool narrative. We are now the team that manages to discover new ways to implode at the crucial moment, such as the first ten minutes of a cup final, even more so than "unlucky but plucky" Liverpool. We are equally capable of snatching victory from the jaws of defeat and vice versa. With a trophy on the sideboard, we are no longer "specialists in failure" or "weak-willed chokers", but we're clearly combustible and thus not the stuff of champions, no matter how entertaining. We have rediscovered our inner cavalier. In reality, a sober assessment of the data suggests we may be closer to rediscovering some of the roundhead virtues that underpinned Wenger's title-winning squads.

I wrongly predicted that we would finish second in the league, based on the assumption that it would be a low-points year all round. In the end, Citeh regressed to the mean with a typical winners tally of 86, while the runners-up total was a high 84. Overall, this was a season of improvement for us, despite the falling away in the New Year as injuries sapped our attacking dynamism. The key change has been a better balance of attack and defence, which the high-profile slip-ups have distracted from. Wenger is a fan of "efficiency", which can be thought of as maximising the return of points from goals and goals from chances at one end, while minimising the opponent's chances and conversion ratio at the other end. Though the example of Citeh and Liverpool both scoring a ton gives heart to romantics, trying to run up a cricket score is usually not the key to success.

We had the best points per goal ratio among the top four of 1.16, ahead of Chelski on 1.15, Citeh on 0.84 and Liverpool on 0.83. That in itself is not particularly significant - the Spuds had the best ratio at 1.25 and spanking other teams, as Citeh and Liverpool did, inevitably forces the ratio down. It becomes significant when it is allied with a low goals against tally. If you win most games 1-0, your ratio will approach an ideal 3. Our obvious problem was in conceding 41 goals, though this obscures two very different types of performance. Across 31 games we conceded a total of 14, but in the other 7 games we conceded a total of 27, with 17 coming in only three matches: away at Citeh, Liverpool and Chelski. If we'd lost those three games by a single goal each, our points tally would have been the same but we'd have had the best goals against record in the league, matching Mourinho's little pony.

Looking forward, our challenge in attack is to improve our conversion ratio of chances to goals from 13% to nearer 18%. Ozil, Cazorla and Wilshere can certainly create sumptuous chances, but they need to emulate Ramsey in scoring more. Our defensive challenge is to cut out the heavy defeats against title-challengers, which paradoxically requires a more clinical striker. We've shipped goals when the full-backs have had to chase the game, leaving the centre-backs exposed. A more cautious approach, certainly away from home, would keep us in the game longer. A top-quality striker, who can both outpace defenders and create his own chances, might snaffle a point or three in those key games and perhaps bridge the gap to the title. This might appear hard on Giroud, but having a substitute striker of his quality on the bench (technically accomplished if insufficiently mobile), and the option it gives to play two-up front if chasing the game, is exactly the sort of squad strength that we now need if we are to improve further. Which brings us to the matter of the moolah ...

According to The Economist, money matters more than managers, with relative spending on player wages accounting for 55% of the variation in final league points. This is about as surprising as the sylvan preferences of bears, but the detail is interesting in a couple of respects. The overall distribution shows diminishing returns above 300% (dominated by Citeh and Chelski), with points flattening out to the 85-90 range, which reflects the overall competitiveness of the league. While out-spending everyone else has clearly worked since the arrival of Abramovich in 2003, and is unlikely to be restrained by UEFA's Financial Fair Play rules, the chart also shows that coming first is still achievable with a relative spend in the region of 250%.

Spending on wages at Arsenal, relative to the average spending in the league, has been steadily increasing. Though this hasn't yet improved the final position, it has improved the final points tally. In 2010-11, a 215% spend produced 68 points; in 2011-12, 230% delivered 70 points; in 2012-13, 246% delivered 73 points; and this season saw 253% produce 79 points. These figures suggest a careful balancing act by Arsenal during the stadium development years and perhaps a pointer to better days if increased revenue translates into consistently higher wages and transfers. Certainly, the Ozil investment looked like a statement of intent, specifically to raise spending to the 250-300% range. In retrospect, the bid of £40 million for Luis Suarez looks like part of a clear-eyed strategy, rather than the provocative punt of legend.

It's safe to assume we will be bringing in a new keeper and right-back to replace Fabianski and Sagna, and possibly an extra central defender even if Vermaelen stays. We may also invest in a defensive midfielder as an alternative to Arteta (the rumour is that Wenger wants to reinvent James Milner, which would be eccentric even for Le Professeur). The key signing will be up-front, with Loic Remy and Karim Benzema widely mentioned. The latter is certainly the type of striker we should be targeting, though his availability might ironically depend on Luis Suarez's calculation of Liverpool's chances next season (with more knowing opponents, and without a new defence, they're probably not so good). Snatching Diego Costa from under Chelsea's collective nose would be even better, though for wholly petty reasons.


  1. I am a bit suspicious of this efficiency thing. I think goal difference is the big thing. How about this for an alternative.

    If you're good enough to win the league, you'll have a few games with silly scores (in your favour). You'll lose occasionally, but not by much. So you'll have a good goal difference, and not much "efficiency" - all those wasted 5-0 wins. If you're only good enough to finish third or fourth, you're less likely to have any/as many silly scores in your favour. You're also likely to lose a few more times.

    An additional risk of the almost-but-not-quite-there contenders is that you might try to win against top sides, and risk a catastrophe. And if you're 2-0 down, you won't just go for damage limitation. So a reasonable club with no aspiration but to stay up might have a reasonable goal difference (Swansea/Southampton). A side with ideas above their station (S****) might have a worse goal difference than their position suggests.

    "Efficiency" could just be a nice way of saying you rarely win big, probably because you're not that good. Looking at the table, goal difference looks the big thing.

  2. Goal difference obviously correlates broadly with league position, particularly at the top and bottom of the table, but what is interesting are the variances from that tendency. For example, Crystal Palace went from 20th place and a GD of -15 at the end of November to 11th place and a GD of -15 by May That's why Tony Pulis won Manager of the Season. He made them more efficient.

    Wenger publicly recommended Pulis for the award, despite their fractious history, because he recognised the ability to produce that extra efficiency. For a team near the bottom, this can add 9 places. For a team near the top, the swing is probably only 1 or 2 places, but that can be dramatic. Palace finished 15 points above 20th. We finished 7 points off 1st.