Posts

The (data) revolution in fielding

While batting and bowling have had quantifiable numbers associated with them, in order to help judge success or failure, fielding has remained untouched until very recently. Jonathan Liew from The Telegraph looks at how England are using a similar model as CricViz to monitor fielders’ performance.

Phil Oliver runs the stats company CricViz, which provides advanced data analysis to fans and broadcasters using a model Leamon helped develop. He believes that the next step is wearable technology, that would allow player’s movements to be tracked to the nearest pixel. Used in conjunction with a Hawkeye-style ball-tracking tool, it would allow us to settle debates that can now only be conjectured. Was that a genuine chance, or just out of his reach? Could the fielder have got to that ball quicker? And who really is the best fielder in the world? (According to CricViz data, Ben Stokes was the most effective fielder at last year’s World T20 by some distance.)

ANALYSING FIELDING IMPACT IN THE BIG BASH LEAGUE

At CricViz we record and rank every fielding moment of note by prescribing a run value and a difficulty to each incident. A full explanation of our methodology for our fielding rankings is included in our PlayViz definition.

Collecting this data allows us to rank every team and every player by adding together all of the separate fielding incidents. At the end of the regular Big Bash League season we’ve had a look at how the leaderboards for players and teams stack up.

Team Leaderboard

The team fielding impact leaderboard is very similar to the final league standings, indicating the importance of fielding to winning matches. Indeed, 23 of the 32 matches in the regular league were won by the team who scored a higher fielding impact rating in that match.

Interestingly Perth Scorchers’ dropped catches does not correlate with their fielding impact as clearly as the other seven teams. This is largely because they dropped four ‘easy’ chances ranked as likely to have been taken at least 70% of the time—harming their impact—while also dropping four ‘hard’ chances ranked as likely to have been taken less than 30% of the time—adding to the dropped catches count.

Fielding RankingTeamFielding ImpactDropped CatchesLeague Position
1Perth Scorchers+25.4091
2Brisbane Heat+20.0952
3Sydney Sixers+17.4563
4Melbourne Renegades+6.0545
5Melbourne Stars-13.8044
6Hobart Hurricanes-23.45107
7Sydney Thunder-31.15128
8Adelaide Strikers-31.5096

Player Leaderboard

To an extent the individual rankings are limited by sample size with no one player having played more than eight matches and each player having only a dozen fielding incidents at most. To facilitate for this, the individual rankings only include those players to have been involved in at least three fielding incidents. While this is still a small number it gives some indication of quality.

Seven of the top ten players did not record a single negative impact. Kurtis Patterson is the only top ten fielder who is not listed thanks largely to a single significant fielding moment (Chris Jordan, Ashton Agar, David Hussey, Clive Rose, Liam O’Connor and Ashton Turner all took a catch of at least +5 impact; George Bailey, Alex Ross and Marcus Harris all made a run out with a direct hit of at least +5 impact).

PlayerTeamFielding Impact
Chris JordanAdelaide Strikers+14.30
Ashton AgarPerth Scorchers+13.65
David HusseyMelbourne Stars+11.65
Alex RossBrisbane Heat+11.20
Clive RoseHobart Hurricanes+10.90
Liam O'ConnorAdelaide Strikers+10.30
Kurtis PattersonSydney Thunder+9.35
George BaileyHobart Hurricanes+8.90
Marcus HarrisMelbourne Renegades+8.35
Ashton TurnerPerth Scorchers+8.00

 

Lowest Ranked Fielders

Perhaps unsurprisingly three wicket-keepers, who are involved in the most incidents, populate the lowest ten ranked fielders, and perhaps less surprisingly still, Ben Dunk, Adelaide Strikers’ part-time keeper who was replaced behind the stumps by Tim Ludeman half way through the season, has the lowest ranking of the lot.

Some of the lowest ranked players are largely there due to one or two basic errors (Kieron Pollard, Luke Wright, Ian Bell) which are punished significantly because they should, and most of the time would, have been taken. Others are there due to a handful or errors (Ben Dunk, Peter Nevill and Travis Head) or at times a litany of errors (Andrew Tye, Ben Laughlin, Cameron Boyce and Brad Haddin). Despite having taken an excellent catch in Match 27, Laughlin makes the lowest ranked list having been involved in five negative incidents.

PlayerTeamFielding Impact
Ben DunkAdelaide Strikers-20.10
Andrew TyePerth Scorchers-16.80
Ben LaughlinAdelaide Strikers-15.30
Kieron PollardAdelaide Strikers-14.25
Luke WrightMelbourne Stars-14.10
Cameron BoyceHobart Hurricanes-13.75
Ian BellPerth Scorchers-13.25
Brad HaddinSydney Sixers-13.10
Peter NevillMelbourne Renegades-11.75
Travis HeadAdelaide Strikers-11.60

Freddie Wilde is an Editor and Analyst at CricViz. @fwildecricket. 

ENGLAND CATCHING ON

When the England squad headed to Spain for a pre-series training camp it was derided as a holiday by the Australians. Some form of team bonding was required by a team that had only just been introduced to new coach Trevor Bayliss, but it became clear that their Almeria trip was far more than a jaunt in the sun.

In his post-match interviews at Trent Bridge Alastair Cook placed heavy emphasis on the fielding preparation done in Spain. The slip cordon became settled and hard work was done, with the captain keen to point out how much catching practice was conducted.

The rewards are clear – England have saved more runs through their fielding than Australia. CricViz measures the fielding actions of both teams in each game, producing a run value that their fielding has had on the opposition’s score. The use of projected averages and a detailed rating system allows the accurate measurement of fielding impact.

Australia recorded negative fielding scores in each of the first four Tests, combining to produce a total of -124. England saved runs in three of these four Tests, heading to The Oval with a fielding score of +67.

Fielding impact (runs)EnglandAustralia
Cardiff84-26
Lord's-133-58
Edgbaston41-2
Trent Bridge75-38
Total67-124

The worst fielding score of the series so far was England’s -133 at Lord’s, part of a performance where nearly everything went wrong. England showed they could bounce back better from a nightmare performance than Australia, and this was especially true in their fielding.

They dropped just one chance at Edgbaston – a difficult opportunity that flew high through the slips – and were flawless in the first innings at Trent Bridge. Not only was every catch opportunity taken, but no ground fielding errors were recorded in Australia’s 111-ball procession.

With the urn within reach chances were spilled in the second innings, but the work done in Spain was evident. Ben Stokes and Joe Root pulled off memorable diving efforts and whilst Steven Smith did something similar for the tourists, it was an act of defiance that did not represent the team’s fielding standards.

1ST ASHES TEST ANALYSIS

FIELDING COMPARISON

With two well-matched sides, each batting and bowling well, to a large degree the deciding factor in the series opener was the quality of their fielding in the first innings. In a game where both sides got a number of half-chances, England were sharp and clung onto theirs, Australia spilt a few and suffered as a result.   The difference between the impact of the two sides’ fielding in the first innings was 113 runs, almost the entire 1st innings lead that gave England control of the match.

Eng Fielding ScoresAus Fielding Scores
1st Innings51-62
2nd Innings3336

At the end of Australia’s first innings WinViz had England at nearly 70% to win the match.

Take away the 113 runs between the teams’ fielding and the situation would have been different. England would still have had a small edge – Australia still had to bat last on a wearing wicket – but it would have been far more evenly poised contest.

1st-Test-CricViz-Analysis-WinViz-2

AUSTRALIA’S AGGRESSION AGAINST SPIN

Australia pursued a policy of aggression against the English spinners, but in doing so lost 7 wickets for 158, including 4 key top order wickets to Moeen Ali. Australia’s record against spin overseas has been poor in recent years, and Ali was the bowler against whom they underperformed most in this match.

Test Avg Overseas – since 2010
SpinPace
Aus30.732.3
Eng35.230.8

From the Hawkeye data, BatViz predicts that an average Test batsman would have attacked 39% of the balls bowled by English spinners in Cardiff. The Australians attacked almost exactly half. On this occasion, the strategy hurt them considerably.   With long periods when there was little assistance for the spinners from the pitch, BatViz estimates that an average Test side should have averaged 45 against spin in this match, but instead Australia lost their wickets at 22.6.

Australians v Spin in Cardiff
BatViz PredictionActual
Batting Avg45.122.6
Attacking %39%50%

DIFFERENT APPROACHES FROM THE TWO SETS OF BOWLERS

CricViz’s analysis of the two pace attacks shows that while both sides bowled well in Cardiff, they did so in slightly different ways. Australia bowled slightly quicker, and swung the ball more in the air.

England in contrast, were able to get more movement off the wicket (often through the use of cutters) and were far more accurate. Australia were able to induce slightly more mistakes from the batsman, England did so in more dangerous areas.

FLUCTUATIONS IN THE CONDITIONS

With little pace or life in the surface, the pitch became more of a new ball wicket as the match went on.

BatViz Predicted Average by phase of innings
BallsInn 1Inn 2Inn 3Inn 4
5027.522.233.521.7
10027.632.326.734.3
15024.338.639.135.3
20031.545.641.735.0
25033.337.136.040.7
30032.439.734.149.5
35035.649.436.936.8
40037.633.231.838.5
48031.841.825.3
53028.623.5

On the first day, under cloudy skies, the ball swung for most of the day, and batting although slightly easier after the first two hours, remained difficult all day. As you can see from the graph, England’s new ball spells were more potent, but as the ball stopped swinging they were unable to sustain the threat to the batsman that Australia had in more helpful conditions on Day 1.

PLAYVIZ

This was a high quality encounter. An excellent Australian side buoyed by recent successes, and a good, young England side playing in their home conditions. As we can see from the PlayViz output, the general standard of play was very high.

1st-Test-CricViz-Analysis-PlayViz-1

Over the course of the match, England’s batting was 79 runs better than an average Test side’s under the same conditions, their bowling 81 runs better and the quality of their fielding was worth another 84 runs.

Australia’s bowlers were outstanding, 150 runs better than a typical attack, but they were let down by their fielding, particularly in the first innings. The Australian batting, whilst 17 runs better than a par Test side, was also down on their usual performance levels.