Here’s some brief notes written ahead of the first Test. I really should have put this up before the Test started. Anyway:
I give England only a 31% chance in the first Test. The betting markets say 39%. Why the difference? The toss is vital and England’s batting isn’t at full strength.
- Batting first is key. SL are W7 L1 D1 batting first, W3 L4 D0 batting second recently. Batting first is worth 148 runs (runs per wicket by innings over the last 10 years: 40, 28, 29, 26). A 400 pitch becomes a 280 one after the successful tossers have had their fun with it.
- Note spin is no good in first innings (average 42, SR 77). If you field first and get nowhere in 20 overs, you are in very deep trouble.
- England have a lot of right handers. A tasty matchup for a leg spinner or SLA bowler. There are two in the Sri Lanka squad: Lasith Embuldeniya averages five wickets per FC game, PWH de Silva is more an all rounder who averages two per game. Embuldeniya averages 40 after seven Tests, but with a FC average of 25 in Sri Lankan conditions, he has a great opportunity. Surprised to see Embuldeniya’s odds 25-1 for Man of the Match. Oh, and he’s Sri Lanka’s leading wicket taker over the last two years.
- On the topic of Sri Lankan FC averages, there’s a gulf between Test Cricket and the Sri Lanka Premier League Tier A. It’s hard to estimate because there are few (if any) overseas players for calibration, but I make the increase in bowling average 70%: a 25 average in Tier A translates to a Test average of 43. Here’s the expected averages for Sri Lanka’s attack:
- Away teams pick too many spinners (over the last ten years away spinners average 35 at Galle) likely because teams pick more spinners than are Test standard. The relevant decision is “who will do better, our third spinner or our first change pace bowler”?
- In England’s case they that’s not a question of spinning ability, more the balance of the side. With Ali unavailable, England don’t have the batting depth to pick a third specialist spinner. Expect Curran+Bess+Leach+Two Pacers+Root. Sri Lanka will know this, so have an incentive to prepare a spinning pitch and nullify England’s pace attack. Unclear what the pitch will be like as has to be good enough to take back-to-back Tests.
- Curran and Bess may not offer enough in either batting or bowling to balance the team. Maybe in a couple of years, but today England look beatable.
- Put all that together, England have the better bowlers, but the toss is so important that it’s a great leveller. Win the toss, bat, win the game.
Changing conditions
I hadn’t noticed this change – it used to be that the 2nd innings was the time to bat in Sri Lanka. Now it’s the 1st innings. See below the difference in runs per wicket from batting first/third versus second/fourth. A big advantage to winning the toss and batting.
Why should that change happen? Different groundsmen? Different grass? Playing at a different time of year? Either way it shows the importance of “live” queries feeding models rather than fixed assumptions.
- PS. Reflecting after the first day’s play I need to think about specific matchups. Bairstow and Root are good against spin, even when it turns away from them.
- PPS. There’s a lot of Test series happening right now – will December/January become the annual window of international red ball cricket?
- PPPS. The comments about the importance of the toss look silly when spin took 6-85 in the first innings. Was I wrong or were Sri Lanka’s batsmen wrong? Hard to gauge without xW data.
Red Ball Data 