This article is about the difference KW makes to a BCs ATG ODI XI.
Kane Williamson may have just turned 25, but he's already a certain pick for the No. 3 position in an All Time Black Caps ODI XI. This article uses an online ODI cricket simulator to look at the effect that the coming of Williamson has had statistically on teams he's been in, and what this could mean at the gambling houses.
In being elevated to the Black Caps' All Time ODI XI at No. 3, Williamson replaced Andrew Jones, who had been there for about 20 years. Jones was not only our second best ODI No. 3 but he was second behind only Martin Crowe in their era. He averaged 35.69 with the bat at a strike rate of 57.9, scoring 25 fifties in only 87 matches.
Comparing Jones's stats to his contemporaries puts him much closer to Martin Crowe than it does to the other mainstays of the batting lineup such as John Wright, Mark Greatbatch and Ken Rutherford. Jones was a very, very good player.
But how much better than Jones is Williamson? One way to find out is to use ODI CricSim. In this simulation, I went into custom mode and set two almost identical teams against each other. Choosing the preset "New Zealand ATG" side from the NEW ZEALAND drop-down menu, I replaced Williamson with Jones in one team and left the other as it was. Then I hit the UltraTurbo checkbox and left it to run for a bit.
The simulation ran until the stronger team had won 14,000 matches, the first 7,000 wins from batting first and the other from batting second. The result was that the Kane Williamson XI defeated the Andrew Jones XI by 14000-10644 ($1.77 to $2.33, 176 ties). So this improvement by one player alone is worth about 56c to a bet on New Zealand.
This piece of data could be of considerable use to a person betting on BetFair. Should Williamson get injured and miss a match, it's likely that his replacement will herald a significantly lower chance of the Black Caps winning. Even if his replacement was as good as Jones - which is statistically improbable - an otherwise identical team would be around 56c more likely to win. Williamson is that good.
Of further interest is that Williamson is a sufferer of the Martin Crowe effect, in which a young player too good for the domestic leagues but not up to international standard is thrust into the international side anyway, meaning that their overall career stats do not reflect how good the player was for the bulk of his career, as his apprenticeship is factored in. Should I have used Williamson's past 2 years average of 59.85 instead of his career average of 48.02, the margin would have been greater still.
If Williamson did get injured and get replaced by an average player, the BetFair punter has other things to be wary of. In the Williamson-Jones ATG simulation, Williamson ended up with a $4.51 fair value to top score in any given innings. The corresponding figure for Jones was $6.25. This is really what I mean when I talk about the Kane Williamson effect. Martin Guptill opening the batting - in teams identical save for the choice of Jones or Williamson - is paying $5.30 in a team with Williamson in it and only $5.03 in a team with Jones in it. Martin Crowe, coming immediately after Williamson at 4, is paying $6.56 in a team with Williamson in it and only $6.00 with Jones.
This means that, should Williamson miss a match for some reason, fair value for bets on other batsmen top scoring could vary by 50c or more.
In the simulation referenced in this article, Williamson and Jones faced roughly the same number of balls (Williamson 1,087,959, Jones 1,163,021). But there was a considerable difference in total runs scored from these balls (Williamson 895,134, Jones 719,320). This difference in batting approach is reflected in the economy rates of the opposition bowlers.
All five of the opposition bowlers had a BowlEcon 0.2 runs per over higher when bowling to a team that had Williamson in it (their strike rates were roughly identical). Over 50 overs, this is in the area of ten runs per match. Most of this will be the difference in averages between Williamson and Jones, but some of it will be from the handful of extra balls at the end of each innings from the fact that the average Williamson innings took up fewer balls.
Naturally, this gap in strike rate is reflected in the fair value odds of either player scoring a century. Jones, who famously never scored an ODI century despite passing 50 no fewer than 25 times, is paying $62 to ton up. Considering that he played 87 matches, and that the bowlers Jones faced had much lower economy rates in general than Corey Anderson, who is used in this simulation, it's perhaps not unlikely that Jones would never score an ODI hundred.
Williamson, on the other hand, is paying $16.89. So he is roughly four times more likely to score a century than Jones. In real life, Williamson has scored 7 centuries from 85 matches, which works out to $12.17. Considering the bowling in this ATG simulation is far better than the average quality of bowling Williamson has faced in his career, this figure is therefore fairly likely to be accurate.
