By now readers are familiar with my research on the importance of winning streaks as a predictor of Arsenal winning the league (read my first and second blog on the subject). So after last weekend the streak now stands at 6-wins and the league table now reads:
| Position | 2nd |
| Games played | 8 |
| Wins | 6 |
| Loss | 1 |
| Draws | 1 |
| Points | 19 |
| Points Per Game | 2.38 |
So despite the hysteria and portents of doom by the emotionally driven majority in the fanbase, after a slow start the club has rapidly made up ground on a now stumbling Manchester City and is only second place in the league by dint of goal difference.
Historically Arsenal have won the title by streaking; 13 games in 2001/02, 10 games in 1997/98 and 9 games in 2003/04. Should the current winning streak continue, I concluded it would be a massive indicator of a serious title challenge. Currently the data is not conclusive and I would urge fans to not pop any champagne bottles, at least not just yet.
Why my caution? Because history shows, since last winning a title in 2003/04, the club has had a habit of starting very well and fading down the homestretch. This is evident in the data for Points Per Games (ppg) mid-season and full-season:
| Mid Season | Full Season | |
| 2004-05 | 2.16 | 2.18 |
| 2005-06 | 1.74 | 1.76 |
| 2006-07 | 2.06 | 2.04 |
| 2007-08 | 2.35 | 2.18 |
| 2008-09 | 1.75 | 1.89 |
| 2009-10 | 2.16 | 1.97 |
| 2010-11 | 1.89 | 1.79 |
| 2011-12 | 1.89 | 1.84 |
| 2012-13 | 1.74 | 1.92 |
| 2013-14 | 2.21 | 2.08 |
| 2014-15 | 1.74 | 1.97 |
| 2015-16 | 2.05 | 1.87 |
| Average | 1.98 | 1.96 |
In raw numbers it may seem to be a mere 2 basis points difference but proportionately it is a 10% negative swing. Mathematically that is significant reversal; from an average of 38 points in the first half of the season to 36 points in the second, gives a season average total of 74 points. Over the same period the winning team in the EPL averaged 2.30 ppg or a season total of 87 points, a whopping 13 point superiority over Arsenal.
Despite years of wilful misinformation by the mainstream media and so-called Arsenal bloggers, in their desire to have Arsene Wenger sacked, the primary reason why the club faded over the stretch was not having sufficient resources to build a squad of with the requisite quality and depth, given the priority of having to pay for a new stadium. I have spent several blogs doing the pre-Emirates vs post-Emirates comparison to show Wenger was both a genius for winning titles at Highbury as well keeping Arsenal competitive post-Emirates despite being massively outspent by his main rivals. The data is unchallenged and bears no repeating.
But despite the spending handicap over the past 12 years, the club has occasionally out-competed the entire league at the start of season; sitting at the top midway but eventually fading. However final league position has progressively improved over the last three years, from 4th in 12/13 to 2nd in 15/16. The graph below illustrates.
Average Points Per Game (PPG)
But the number I found to be significant, streak or no-streak, is the average ppg. After 8 games AFC is literally steaming at 2.38 ppg. Continue at this rate and the club will approach averages last seen in the Invincible year, which was a steady 2.37 ppg both midway and full season. Since those heady days, the next best was the average 2.35 ppg midway 2007-08, the year when our title challenge was literally smashed with the multiple fracture of Eduardo’s leg at Birmingham City in March 2008.
Average PPG – Premier League Champions
| Mid Season | Full Season | |
| 2004-05 | 2.42 | 2.50 |
| 2005-06 | 2.75 | 2.38 |
| 2006-07 | 2.47 | 2.34 |
| 2007-08 | 2.18 | 2.26 |
| 2008-09 | 2.25 | 2.37 |
| 2009-10 | 2.25 | 2.26 |
| 2010-11 | 2.11 | 2.11 |
| 2011-12 | 2.37 | 2.34 |
| 2012-13 | 2.45 | 2.34 |
| 2013-14 | 2.16 | 2.25 |
| 2014-15 | 2.42 | 2.29 |
| 2015-16 | 2.05 | 2.13 |
| Average | 2.32 | 2.30 |
As the table above illustrates, since 2004 the average winner of the PL title has averaged 2.32 and 2.30 ppg midway and full-season respectively. Interestingly the most dominant mid season performance was Chelsea’s in 2005-06, a year they suffered only one loss, coming very close to matching the Invincibles. The least dominant mid-season performance was last year’s, by Leicester, at 2.05 ppg. They were however able to improve their average point-haul by 39% in the second half of the season. In contrast over the 12 year period, most clubs set out to dominate in the first half of the season and suffer some tailing off in the 2nd. This happened 7 out of 12 times. But surely anything above the mid-season 2.32 ppg average is statistically significant.
In my opinion it is too early to come to any conclusions about the current streak. We are just half way into the first half of the season with ten more games to go. Clearly some patterns are beginning to show which we can all track and measure. I leave it to you dear readers to speculate.
Positively Arsenal 
The difference between this streak and other recent ones is the depth of this current squad. We’re packed in midfield, across the back, at goalkeeper, up front, everywhere.
Last season, losing Ramsey so early would have killed us. This season we have Iwobi taking up the slack and we have the Ox waiting to take his spot should he need replacing. In other words, we’re stronger than we have been for a while.
LikeLiked by 6 people
Need to sleep but I was snared by the title and the painted lady!
Another wonderfully intriguing round from Shotta’s “one pop”. As a football fan, instinctively I don’t want any part of the beautiful game to be modelled and/or predictable. Having said this, I think something is different about Arsenal this season and it is very reassuring that Shotta’s hard work is in sync with my musings.
Has anyone noticed that in the last few years, newly promoted teams are no longer cannon fodder? It seems tactical awareness and performance levels have improved significantly leading to shrinking differentials between the top, mid-table and bottom teams.
Excluding the winning formula that Leicester “lucked” upon last season, all top-tier teams seemed to struggle with teams many positions down the table. I recollect lots of very unpredictable outcomes to what should have been very predictable games. Take Spurs (the gift that just keeps on giving), 2nd in the league, losing to 10 man, relegated Newcastle. Could we be witnessing the beginning of a new era of very modest full season PPG’s for the future Premier League Championship winners?
LikeLiked by 2 people
So despite the hysteria and portents of doom by the emotionally driven majority in the fanbase, after a slow start the club has rapidly…
Oh man I was looking forward to reading the article until that pile of arrogant smug after the event tosh!
Football supporting is 90% emotion, or it is for me anyway. Suggesting you’re above all that and mocking others for being “emotional” or pissed off at the disarray we were in the lack of signings (again, but of course it was all part of the master plan!) and the debacle that was the Liverpool game (let’s be honest we were pretty ropey until after the PSG game) is a bit naff. And given we’ve achieved nothing yet is a bit premature anyway.
Realists are usually found on le Grove!
I liked the rest though!
Yours
Emotionally charged of Wembley
LikeLike
Long may we streak on! Cheers Shotts for collating the data.
LikeLiked by 2 people
Good morning Shotts and thank you for another meticulous scoop and analysis of the data.
While the figures this year are good it may also be a further benefit that our PL games have included the reigning champions away, our recent nemesis Chelsea, and the massively improved Liverpool. In previous seasons points away to the reigning title holders and the team from SW6 would have been a cause for celebration in themselves.
Boro to come so fingers crossed that we move to 2.44444444 by Saturday pm and 2.5 on the coach back from the SoL.
LikeLike
I was snared first by the title, then by the painted lady! Another wonderfully intriguing round fired from Shotta’s “one pop”. As a football fan, instinctively I don’t want any part of the beautiful game to be modelled and/or predictable. Having said this, I think something is different about Arsenal this season and it is very reassuring that Shotta’s hard work is in sync with my musings.
Has anyone noticed that in the last few years, newly promoted teams are no longer cannon fodder? It seems tactical awareness and performance levels have improved significantly leading to shrinking differentials between the top, mid-table and bottom teams.
Excluding the winning formula that Leicester “lucked” upon last season, all top-tier teams seemed to struggle with teams many positions down the table. I recollect lots of very unpredictable outcomes to what should have been very predictable games. Take Spurs (the gift that just keeps on giving), 2nd in the league, losing to 10 man, relegated Newcastle. Could we be witnessing the beginning of a new era of very modest full season PPG’s for the future Premier League Championship winners?
LikeLiked by 1 person
I have just noticed you in the pending file Arsical – thank you for your patience.
LikeLike
Dexter – how did you get in ??
LikeLike
MAllen: “one pop”? As a person obviously familiar with Jamaica’s rude bwoy culture that is a terrible diss. You know a real ‘shotta” only profiles with a high caliber automatic pistol or a machine gun. Only mine is filled with data. Lololol.
LikeLiked by 1 person
The One Boro website is not optimistic this morning;
http://www.oneboro.co.uk/forum/showthread.php?tid=8347
“Juninho9 Prediction: All the MD threads I’ve done have resulted in draws. Would give an arm to have any attacking intent in the game, let alone manage to salvage a point or a clean sheet. But my head says 3-0 Arsenal, if we’re lucky.”
LikeLike
I let Dex in Andrew. I always love to share dissenting opinions and Dexter and I have been on other sides of the debate for years. The fact that fans are emotional is the leverage used by the bastards in the media and many so-called Arsenal blogs to mislead sensible people into believing Arsene should be sacked, contrary to the data. Rather than doing the harder task of appealing to the left-side of our brain, they prey on our weakness. I wish to remind Dex and those like-minded that many used to excuse terrace violence on “emotional” fans. I rest my case.
LikeLiked by 5 people
Dexter playing the old” stop reminding me I acted like a spoilt toddler” get out of jail card.
LikeLiked by 2 people
Shotta_gooner: merely an attempt to introduce levity and to be complimentary in a cryptic but familiar way. Consider this perspective – anyone intimately familiar with the one pop is definitely “foundation”. Creator, not follower.
Plus, in many of your blog posts you take careful aim and fire one very precise round directly to the heart of the matter. One round is all you seem to need. One pop is apt and yours is irrefutably effective!
LikeLiked by 2 people
Simply brilliant
LikeLike
“As the table above illustrates, since 2004 the average winner of the PL title has averaged 2.32 and 2.42 ppg midway and full-season respectively.”
shouldn’t 2.42 be 2.30 ? your argument seems to fall flat though in this case, and that’s even before we consider using median instead of average to remove outlying seasons
LikeLike
Great stuff Shotts,that ‘thanks for the memories Arsene’ banner must look like the Count of Monte Christo’s vest by now,I’m off on my travels tonight so-top of the table & quarter finals of the league cup when I get back please lads..
LikeLiked by 1 person
most seasons we’re being undone by terrible refs leading to many injuries
this season looks better so far, but moss recently and dean in the next game might start the old trend
if they decide to screw us again – naturally the second half results will suffer
LikeLiked by 2 people
MAllen @12:18pm I feel you my friend. As you probably observed, to solicit readership of this very dry subject I chose a very provocative title and a somewhat salacious picture. Thanks for the compliments. My Jamaican forebears, who resisted slavery, may have had their uprisings drowned in blood but they took very careful aim at the enemy.
LikeLike
Mel: Thanks for the kind words. Is your twitter account protected? I tried to DM you this blog without success.
LikeLike
Shotta. Mel gets an Email of every blog.He is a subscriber, Among other things.
LikeLike
Oleg Ych: Do you calculate mean averages different from Excel? I could have calculated the median but I have no intention of losing my readers by trying to show how many amazing stats I can generate. That is the weakness of all those nerdy stat blogs that dissappear as quickly as they appear (no readership).
LikeLike
Shotta: http://prntscr.com/cx3gaa not sure where does the 2.42 come from
LikeLike
A very interesting read Shotta.
I have always tried to rail against stats – or at least the over-bearing reliance on them – and I regularly used to rail against one commentator in particular on ACLF for his love of stats which he’d dish out to ‘prove’ whatever point he was attempting to make at the time.
I think stats can be useful in determining (or at least indicating) trend and that’s pretty much how Shotta’s used them here. But our midweek fixture saw A Bulgarian Team with a staggeringly high possession stat to sufficient to confound an army of statisticians. And one of our best wins against Man City in recent years saw us end the game with what you might have assumed to be perilously low possession stats (was it as low as 39% a couple of seasons ago?).
And in some ways I have sympathy with what looks like the hacked account of Dexter; football is and should remain a game that plays on the emotions rather than the abacus. But just as an over-reliance on stats can sometimes prove unhelpful (I’m not leveling that at you Shotta), so too can the release and reliance on unbridled emotion leave you looking incredibly foolish at times.
At the start of every season my unbridled emotions see me confidently expecting Arsenal to sweep all before them. Optimistic – just a tad? Foolish – very probably? But I’d rather be that way than join the weird ‘realists’ of le Grove whose heads tell them that because our title charges usually tend to be derailed by an annual injury crisis or two, that that will always be the case. And stats alone suggest we can’t always be that unlucky. The odds-defying success of Leicester City suggests there’s some truth in that.
So when we look at the stats Shots provides and combine that with what Mel’s waters are likely telling us then it’s a pretty safe bet that we are in with more than a decent shout this season. Can we measure squad depth using stats? The bursting into form Theo? The sudden emergence of Ozil as a free-scoring mid-fielder?
No we can’t; but our emotions tell us these developments are all as potentially significant as any unbeaten run.
LikeLiked by 2 people
Oleg: My apologies. The error is in my narrative below and not in the table which I copied from an Excel file that I inputted the data. Printer’s devil they used to say in the old days. I will fix it when I get home from work.
LikeLiked by 1 person
TEAM NEWS: SANTI, RAMSEY AND WELBECK
Arsène Wenger has given an update on our latest team news ahead of Saturday’s game against Middlesbrough.
on the latest team news…
Well, from Wednesday night it looks like we have no big problems. The only uncertainty I have is Santi Cazorla, who got a kick. We’ll see how he is today but we have nobody coming back. Of course we lose Xhaka because he’s suspended and everybody else should be available.
on Ramsey and Giroud…
They are progressing well. Olivier Giroud could maybe be involved on Tuesday and certainly will need one more week sharpening up. He could be available but we want him really to be 100 per cent fit before we bring him back.
on Ramsey’s progression…
Ramsey is progressing well and I believe fitness-wise he is 90 per cent there. His potential return could be not against Reading on Tuesday, but maybe Sunderland.
on Welbeck..
He could be back in January. After that let’s not forget he was already out for a year, so we might have to be patient. To get him back to a really competitive level… it will certainly take two months of competition to get back to his best.
on what stage he is at…
No [he is not out running yet], he will be starting to run in two weeks.
Copyright 2016 The Arsenal Football Club plc. Permission to use quotations from this article is granted subject to appropriate credit being given to http://www.arsenal.com as the source
Read more at http://www.arsenal.com/news/news-archive/20161021/team-news-santi-ramsey-and-welbeck#Jmox19EDymZcPlLy.99
LikeLike
Shotts,I wouldn’t know how to lock a Twitter account! As George says I subscribe to PA anyway so always get the blogs along with the S&M weekly that George forwards me😀
LikeLiked by 2 people
Ozil has given the WOBs a massive kick in the bollocks.
LikeLiked by 1 person
I’ve really enjoyed Shotta’s teasing us with stats on winning streaks. Of course the team that wins the EPL is the one that gets the most points. And winning streaks get lots of points. So it’s not surprising that winning streaks and winning the league go hand in hand.
But, as I’m sure Shotta knows full well, it’s how many you win – not the order in which you win them – that matters. WWWWWLWWWWWL is just as likely to win the league as WWWWWWWWWWLL. In fact, if the penultimate opponents in the sequence happened to be (say) ManC, the first one is preferable.
So, if our current winning streak breaks down prematurely, let’s not get too distressed. Let’s just start another one.
LikeLiked by 3 people
I think you’re all being harsh on Dexter it was obviously a parody comment. He couldn’t possibly expect anyone to take seriously the accusation that we were in disarray or suffered from a lack of signings. Come on guys nothing worse than cracking a gag and no one getting it.
LikeLike
merlot – you’d imagine that it wouldn’t matter but as Shotta point out the facts are that teams who win the league do go on long winning runs rather than win more sporadically.
LikeLiked by 3 people
Yes Stew, perhaps Dexter hasn’t gone to the dark side and is just pulling a Muppet.
Then again………………
LikeLike
I propose “Ozil has given the WOBs a massive kick in the bollocks” as our thought for the day, sound bite, mantra of the week, mantra of the month, logo on a T shirt, salutation with champagne, the first line of a stanza in a poem, and a prayer before bedtime. Need I say more, other than PG you are an inspiration?
LikeLiked by 1 person
I’m watching the CL league game again. I can do as I please, the battle n’strife is away tracing her ancestry over in the Emerald Isle. Sanchez just scored that fuck off goal again….
LikeLike
I was inspired by PG that I forgot what I wanted to express. My gratitude for another great and informative read from the mighty Shotta….
LikeLike
Steww – what I love about stats is that they’re so counter-intuitive. .
“teams who win the league do go on long winning runs” – often true, but not entirely surprising.
“teams who win the league do go on long winning runs rather than win sporadically” – well, there’s a whole PhD thesis for somebody there.
LikeLiked by 2 people
It is not the sequence or the serial, it is the anomaly that validates the purpose of statistics. See the irregularity and spy the hidden truth, or possibly even the future.
LikeLike
Got to go with merlot here.
Looking at the Arsenal and the league champions’ tables, I see nothing there to convince me that first half points haul or point per game is more relevant to winning the league than over the 2nd half. I just cannot see any statistical significance. Sorry.
Ultimately, and I think this is basic arithmetic/logical point, the team with most points wins and the order isn’t important at all. We are enjoying a magnificent streak that I hope last forever but I am wary of how it will affect us when it ends after we’ve based our season’s success on it.
To support merlot’s point again, when this streak ends (as they always do), we’ll just dust ourselves up and start another. It’s what champions do and it what we should do to succeed this season.
I have enjoyed all of Shotta’s statistical articles but this one looks too stretched for my liking. I am going to say a “hell yeah” to our glorious streak continuing tomorrow and beyond!
LikeLiked by 1 person
Now we are talking facts. From the data, winning streaks are clear signal of a team’s superiority over the rest of the league but the most telling statistic is Average Points Per Game. Leicester is an outlier at the low end and so was Chelsea at the top end. An average of 2.32 points per game is title winning consistency. For the benefit of the statisticians out there, I did a t-test on that number and the result was a 95% chance of winning the title.
LikeLike
If a team wins lots of games in a season, it is highly likely that that team would put together some extensive back to back wins. This is perfectly logical but that same team could still fall short of another rival with less back to back wins but more consistent run overall. To put it another way, league champions have long winning streaks because they win lots of games and not the other way round.
Talking about facts and stats, they only exist because “things” happened. Stats/data/facts are created from things that have happened and as more and more different things happen, more stats are created. Stats are nothing but numerical history. The notion of the most consistent team winning the league, however, is not a stat; it is logic, which unlike stats is a cold, unyielding bastard.
Here is to hoping that we are both right. May the winning streak continue and may Arsenal be the most consistent team in the league.
LikeLiked by 2 people
I really appreciate Merlot and Bootoomee’s contributions as it helps us, and me in particular, to better interpret the data, without resorting to bias and emotion. This is an opportunity me to emphasize that this was not really a blog about the streak. Streaking was just a salacious segue to lure readers into discussing my thesis that the important stat to watch going forward is the Average Points per Game. If we can maintain an average of at least 2.32 ppg not just midway but over the entire season, rather than fall off a cliff between January and May, then we will make it to the winner’s enclosure. Most of us in PA are experienced and rational enough to know the current streak won’t last forever and we will go through a nervy patch or patches going forward.
LikeLiked by 1 person
[…] point did his or the team’s overall PPG hit that magic number of 2.32 which I have demonstrated here is the statistically tested number to hit if Arsenal is to have a 95% guaranteed chance of winning […]
LikeLike