When was the Ashes won?

The important old ball

England retained the Ashes by securing a draw in the 3rd match at Old Trafford, and then won the series by winning the 4th match at Chester-le-Street.

An interesting question, however, is what parts of the games did England win? It's hard not to recognise the efforts of Ian Bell, who has been immense for England with the bat, often on quite difficult surfaces. Despite his efforts, I think there has been another factor that has been equally influential. England's pace bowlers have bowled well with the old ball.

For the purpose of this article, I've defined the ball as being new if it is less than 20 overs old, and old if it's more than 20 overs.

Here are the difference between the two sets of pace bowlers:

Average	Australia	England
less than 20 overs	22.78	32.32
more than 20 overs	35.28	28.78

Australia are significantly better with the new ball, but England are better with the old ball. While those 7 runs per wicket doesn't seem like a big difference, it is because there have been just over twice the number of overs bowled with the old ball as with the new ball.

Part of the difference has been that Australia seem to have worked hard to keep the shine on the ball, while England seem to have been trying to get the ball to reverse, and not been so worried about using the first new ball as a weapon. England have done much better with the second new ball than with the first (England average 4 runs per wicket more with the first new ball than with the second), while Australia have been the opposite (Australia average 5 runs per wicket more with the second new ball than with the first). Their bowlers almost seem to have run out of steam once the second ball comes along.

Here's how it works out as averages scores. After 20 overs Australia's average score is about 66/2. England is 58/3. However after 80 overs Australia average 180/9 while England are 225/7. Some of that difference is Graeme Swann, but some of that is just how much more effective England's quick bowlers have been with the older ball.

If we took spin bowlers out of the picture and just looked at the fast bowlers, After 20 overs Australia would still average 66/2 (Swann had been about as effective as the quicks with the new ball). England would have been 47/3. After 80 overs Australia would average 250/9 while England would still have 3 wickets in hand at 211/7.

There's a point, however, where these numbers are ridiculous, as the different batsmen are not equally capable, but they do paint a picture.

However they are ignoring the impact that Swann has on the bowlers that are bowling with him.

I believe that one of the reasons that England's quicks have been so effective with the old ball is the pressure created by Graeme Swann. To quantify this pressure, I looked at a statistic that I called Wicket Assists. A wicket assist is awarded to a bowler who bowls an over directly preceding the one where a wicket fell.

Wicket assists is a statistic that I keep note of when I'm coaching, as it's often a better indicator of how much pressure a bowler is exerting on the batsmen than wickets. An example of this is when I had two bowlers opening the bowling together. One bowled 6 overs, and took 0/8 (included 2 no balls, and a wide that just slipped down the leg side). The other bowled 7 overs and took 6/47. Most of the wickets for the second bowler were a result of the pressure that the first bowler built up. The second one got to keep the ball, and got the plaudits, however the one that I recommended for a higher team was the first bowler.

Graeme Swann has taken 23 wickets in this series, but he has also had 23 wicket assists. That's roughly one wicket every 9 overs while he's at the bowling crease. He creates pressure that allows the quick bowlers to take wickets. Compare his rate to Ashton Agar, who had 2 wickets and 5 assists in the two matches he played. That's a wicket roughly every 24 overs that he's at the crease. Nathan Lyon has been more successful than Agar at exerting pressure. 8 wickets and 6 assists, meaning that Australia pick up a wicket roughly every 11 overs that he's at the crease.

Overall I believe that one of the major differences between the two teams has been what their bowlers have done after 20 overs. England's quick bowlers have been better, and their spinners have been better too. Overall, I believe this is the time that this series has been won.

Penny-pinching Misers

I've recently watched the movie Moneyball. The idea of a killer stat is interesting, and it is interesting if such a thing exists in cricket.

In baseball getting on base is the primary objective when batting. There are times that other things are ideal, but getting on base is always a good thing. Compare this with batting in limited overs cricket. If the team needs 8 of the last 2 balls, and you hit a single, you are making it extremely unlikely for your team to win the game. (Unless you have Andre Adams at the other end, and Graeme Aldridge is bowling). However if you need 2 runs off 7 balls, and Bruce Reid at the other end, a single is a great thing.

Likewise if you are chasing 324, scoring 110 off 155 deliveries is really making it difficult for your team to win. But scoring 101 off 143 when your team is chasing 197 on a tricky pitch is an outstanding effort.

However it is rarely true in recent times that a bowler can bowl 9 or more overs and concede less than 50 runs and it be a bad effort. If a bowler stops batsmen from scoring runs, they are doing their job. Over the past 5 years when teams restrict their opponents to 250 or less they win roughly 2/3 of the matches. To put that in perspective South Africa has the 2nd best winning record in that time, winning about 2/3 of their matches. To put it another way, If Bangladesh managed to keep their opponents to 250 or less every match they would be likely to have the best winning record in Asia.

As a result, bowlers who can regularly keep their runs conceded under 50 are very valuable for a team. Which leads to the question: who are the best at doing it?

Player	Bowled 9 or more overs	Conceded 50 or less	Percentage
Mohammad Hafeez (Pak)	24	22	91.7%
GP Swann (Eng)	38	33	86.8%
DL Vettori (NZ)	53	45	84.9%
J Botha (Afr/SA)	41	34	82.9%
RW Price (Zim)	54	44	81.5%
Shakib Al Hasan (Ban)	68	52	76.5%
KMDN Kulasekara (SL)	37	28	75.7%
M Muralitharan (SL)	41	31	75.6%
B Lee (Aus)	28	21	75.0%
BAW Mendis (SL)	31	23	74.2%
Saeed Ajmal (Pak)	37	27	73.0%
NW Bracken (Aus)	33	24	72.7%
P Utseya (Zim)	54	39	72.2%
DJG Sammy (WI)	28	20	71.4%
Shahid Afridi (Pak)	78	54	69.2%

In that list of 14, there are 10 finger spinners. Some of them don't often bowl that many overs, and only get to bowl that many if things are going well for them, such as Mohammed Hafeez, Nuwan Kulasekera or Darren Sammy, however the other 9 finger spinners are in the 13 most likely bowlers to bowl 9 or more overs.

If we give in to conventional wisdom that we need pace bowlers in a team to give the bowling line up balance, we need to know who are the most reliable pace bowlers. Here are the equivalent numbers for the quick(er) bowlers.

Player	Bowled 9 or more overs	Conceded 50 or less	Percentage
KMDN Kulasekara (SL)	37	28	75.7%
B Lee (Aus)	28	21	75.0%
NW Bracken (Aus)	33	24	72.7%
DJG Sammy (WI)	28	20	71.4%
Z Khan (India)	52	34	65.4%
MG Johnson (Aus)	54	34	63.0%
Mashrafe Mortaza (Ban)	37	23	62.2%
KD Mills (NZ)	35	21	60.0%
SL Malinga (SL)	44	24	54.5%
JM Anderson (Eng)	56	30	53.6%
SCJ Broad (Eng)	53	27	50.9%

An equivalent table for wrist-spin bowlers is rather redundant, as it would only contain Shahid Afridi, as the only wrist spinner to have 20 or more innings where he has bowled 9 or more and conceded 50 or less in the past 5 years.

Perhaps this would suggest that if we were going to pick a stats-based world team for an ODI, our bowlers might be best to have Kulasekara and Lee opening the bowling, Hafeez and Sammy as our all-rounders and Vettori and Swann as our pure spinners. It would be a very difficult line-up to score off indeed.