New regime on Stats Island

What with the Acronym War in full flight, the game’s governance in disarray, umpire referrals being trialled, the Kolpak Era showing encouraging signs of drawing to an unlamented close and Mohammed Asif being a very silly boy indeed, it hasn’t been easy of late to focus on what really matters – runs and, um … oh yes, wickets. Unfortunately, even in the safe, profoundly apolitical arena of Stats Island, I can’t seem to get much, if any, satisfaction.

While watching Wednesday’s Pro40 game between Durham and Somerset, the thrilling sight of Steve Harmison and Liam Plunkett adding 81 for the last wicket - and only just in vain - set me wondering about the highest such stand to win a senior match. Not exactly a left-field query, one would have thought. Certainly not an unnatural one for a fully-qualified anorak.

Infuriatingly, my determined scouring of various Wisdens, Frindalls and Webbers, not to mention sundry reputable databases, all came to naught. Which merely strengthened what has long been a personal bugbear: cricket is not well-served by its statisticians. Or not as well as it might be. And definitely nowhere near as well as the Elias Sports Bureau, Bill James and other likeminded souls serve baseball, the only sport that matches cricket when it comes to being fatally smitten by numbers.

The big difference is situational statistics, which baseball uses for public consumption and cricket does not, though several counties and state sides are currently building up the sort of databases that will render that possible. Does Batsman A consistently fare differently according to the score when he takes guard? Is his temperament such that 80-5 is likely to bring more from him than 280-5? Does Bowler B perform more fruitfully according to ego, ie. whether he is given the new ball or comes on first-change? Is a string of maidens more likely to beget a wicket (a popular theory recently rubbished by hard facts)? The answer to these questions could be extremely revealing and helpful in terms of team selection, batting and bowling order, and tactical approach; they would also be fascinating for spectators and students of the game.

But, again, it is the straightforward stuff where the void gapes widest. Which Test keeper has conceded the most byes in a career? More pertinently, which one has the lowest average of byes per Test? No internet database I know of can satisfy that poser. When Graeme Smith and Neil McKenzie were threatening to bat all day against England at Lord’s earlier this month, a friend texted from the pavilion: what’s the record for the fewest runs to have been scored in a wicketless day? I rummaged and number-crunched the latest Wisden as well as a few databases for an hour and got precisely nowhere, other than to report that, had the openers remained intact until stumps, they would have set a record for Tests in England.

It is not that progress hasn’t been made. Strike-rates, for batsmen as well as bowlers, are now accorded an importance that would have been barely credible to statisticians of 30 years ago - and, given the recession in over-rates, with eminent justification. Yet something niggles: why must the denomination for batsmen be runs per 100 balls? Not least because, in the average ODI or even Test innings, 50 balls constitutes a decent sojourn. The results, 178.53 and the like, strike me as far too big and unwieldy. Why not simply make it runs per ball? A rate of 1.78 would be altogether more digestible.

There have also been some worthwhile if ultimately flawed attempts lately to rank current wicketkeepers in such a manner as to take into account the runs they contributed and those they have given away. By including dropped catches, sadly, subjectivity pollutes. One man’s miss is another’s brave try. Ugliness is firmly in the eyes of the beholder.

Only in time will these shortcomings be addressed and redressed, but the will must be there, and I don’t detect enough of that. Of more immediate concern is the need to find a way of properly measuring the effectiveness of batsmen and bowlers in 50- and 20-over cricket.

Averages have long held sway in the first-class fray, but the right and proper accent on economy and strike-rates means that runs per innings and wicket are no longer sufficient as arbiters of quality (and don’t get me started on the ludicrous homage paid to not-outs). May I therefore propose two new, reasonably comprehensible and complexity-free categories: ESR and SA - Economic Strike Rate and Strike Average.

The first of these excludes averages altogether, and instead marries bowlers' economy- and strike-rates by multiplying them. Since both economy-rate and strike-rate need to be minimised, as the following table testifies, Joel Garner’s ESR in ODIs is superior to those of Wasim Akram and Muttiah Muralitharan, the two highest wicket-takers:

ER SR ESR Joel Garner 3.09 36.5 112.79 Muttiah Muralitharan 3.87 35.4 137.00 Wasim Akram 3.89 36.2 140.82

There is a case for dividing that final figure by the number of overs or games, the better to bring opportunity into the mix - which would only enhance Garner’s standing. That, however, would lead to a preponderance of fractions: never a good idea in terms of punter-friendliness. Baseball’s mathematical appeal doesn’t appear to have suffered unduly for the value it places on numbers to the right of the decimal point, but I still find a batting average of 0.333, for all that broadcasters enunciate it as “three-thirty-three”, a little too diminutive for complete respect.

The second category is more complex. Although I would contend that, in the context of the abbreviated formats, runs per innings is of greater significance than runs per wicket, they should form part of the equation. As, equally so, should strike-rate. The following table reflects this calculation, confining the sample to the highest ODI achievers: those with 2000 runs, who average 35 and whose acquisitions come at 70 runs per 100 balls. The results do no justice, above all, to Shahid Afridi, but then speed is nothing without direction.

Average (A) Strike Rate (SR) SA (A x SR/100)

Michael Hussey 54.92 85.63 47.03 MS Dhoni 48.00 92.18 44.25 Viv Richards 47.00 90.20 42.39 Kevin Pietersen 47.14 86.62 40.83 Zaheer Abbas 47.63 84.80 40.39 Michael Bevan 53.58 74.16 39.73 Lance Klusener 41.10 89.92 38.96 Sachin Tendulkar 44.33 85.49 37.90 Andrew Symonds 40.37 92.78 37.46 Michael Clarke 43.40 80.48 34.93

Again, you could divide the SA by the number of innings, or multiply it by the number of runs. Again, and either way, this would not be a sight for sore or even fresh eyes, much less the brain.

These figures, of course, are anything but flawless. The Economic Strike Rate does not take into account that “Big Bird” Garner was at his beak-dipping peak when there were no Powerplays, scores of 300 were immune and only a gifted/mad few dared attempt a reverse-sweep or over-the-shoulder flip, much less a switch-hit. Similarly, the Strike Average conveniently ignores the fact that batsmen score so much more rapidly now. Of the 12 men who have scored more than 1000 ODI runs at more than 90 per 100 balls, only Ian Smith, Kapil Dev and Viv Richards have not plied their trade within the past three years.

But hey, what the hell. All the evidence insists that it was easier for Don Bradman to pile up runs on 1930s pitches against 1930s fast bowling than for Victor Trumper on turn-of-the-century tracks with flimsy pads and SF Barnes to contend with. A century against the West Indies between 1976 and 1991 was worth a double against virtually any other attack ever assembled. Statistics, particularly those used to analyse sporting achievement, have always been two-dimensional at best. Nothing wrong, though, with throwing down a gauntlet.