Mailbag #3 (November 2021)

Your questions answered!

Thanks again for the latest questions people have asked over the last couple of days for Mailbag #3, so with a few hours to spare this afternoon before Bangla Tigers face Delhi Bulls tonight in the T10 League, here are my thoughts on some of the questions asked…

Srinivas/AD (who asked a broadly similar question): “How do the two new IPL teams construct their squad. Ideal approach?”

I could write an entire article on this and probably will do over the next couple of weeks - there’s no way I could do this question justice without doing so. It’s a fascinating topic for me, because I’ve already helped build a squad from scratch with Birmingham Phoenix in the Hundred and think that this particular project would really suit my skillset as a data-driven recruitment analyst.

The problem that the two new IPL teams have is that they are already at a huge disadvantage with regards to the retention system, which allows the existing teams to pick four players to be retained. In theory at least, this means the two new teams can’t pick from the best 30-odd players in the player pool, which makes their job difficult before it even starts.

In advance of the full article, I’ll just list a few areas where they have to focus on:-

Be aware that the next mega auction will be in minimum 3 years time. You won’t get much of a chance to correct huge mistakes for that entire cycle. Finding players with upside throughout this period of time is very useful, whereas committing to a 35 year old in decline probably isn’t, regardless of how many international caps they have.

Don’t be seduced by big names/reputation. Owners might want this for ‘branding’ or commercial reasons but I’d suggest qualification for the knockout stages would do a more effective job in that area. It’s always worth remembering that any commercial income cannot be used to improve the team either, given the limited budgets - it simply goes back into the owners pockets.

Understand the supply/demand dynamics of the market. Top level death bowlers and genuine all-rounders are scarce resources. So are big-hitting domestic batters. In theory they should attract premium pricing and you might have to pay decent money to secure these type of players. Avoiding overpaying for plentiful resources (e.g. anchor batters, domestic spinners) should be mandatory, but you’d be surprised how many teams make this basic error.

Use data to get a clear picture of the domestic market. For example, very few players without significant IPL experience (unless they are young with big future potential) at SMA/KPL/TNPL level are good enough to be regular starters for IPL franchises. Yet many teams pay big bucks for these lower level players, often influenced by recency bias following a strong domestic campaign.

Avoid recency bias. I think many players who played well at the World Cup (small sample size) or will do so in the next couple of months will get big contracts at the mega auction, when a more robust sample size of data would suggest there will be better options who are either cheaper or will go unsold.

Hire a data-driven recruitment expert. Shameless plug! Teams are pretty reticent to do this, for reasons I cannot particularly work out. If any IPL teams are reading this, feel free to email me.

Virat Kohli (presumably not the real one!): “Great work with the Tigers so far!!! My question, though time-consuming is purely one of recruitment. In terms of upside, we have seen teams like MI invest in players like Suryakumar, Kishan, Bumrah, the Pandyas, only to be released in the auctions. In a country like India, where there are 25+ teams in SMAT etc, how do you go about recruitment to identify gems like MI does”

Thanks for the kind words! I guess I covered some of this in the answer to the previous question, but the best way to find out more about how I’d recruit from the lower levels would be to read my book - I cover this area in complete detail. If you would like to purchase a copy, please email me and I’ll let you know how to do that.

Specifically towards MI, I’d retain Bumrah without even a second thought. He’s the best player in a skillset of scarce resources. Same for Hardik Pandya if he’s fit to bowl on a regular basis, for the same reasons. They’ll have to release at least one of Suryakumar and Kishan and risk not getting them back because one of the new teams will surely be interested in these type of players. This is because I think MI will retain Rohit which I think is a big mistake - in my view his T20 game is in slight decline and that’s a big risk at a high retention price over a three year cycle where it’s basically a given he won’t be released during those three years.

Ram: “What is the best team,venue and position in x1 , role for Benny Howell in IPL if he picked?”

It’s probably not a surprise to most readers that I rate Benny extremely highly. I think he’d be very effective in the IPL, particularly at certain home venues - e.g. Chepauk. He adds balance to a team with batting and bowling ability (he’d probably bat 6/7 there), and to be honest, it blows my mind that many mediocre all-rounders get contracts ahead of him.

Jason: “In recent times, there has been an increased demand for the "pace bowling + finisher" skill set in the IPL. IPL teams tend to overspend on 'known' players such as Stoinis, Christian, Russell, Bravo etc, rather than someone like a Cutting, Howell, Gregory etc, who are pretty decent with the ball and have higher boundary percentages. How do you go about recruiting players who tend to go under the radar even with brilliant performances? And, in IPL, is the scouting to blame for not identifying talent or just the love for big names? Also, how much does recency bias work in favour of auction strategy since BBL happens before every IPl auction, and other great tourneys like the Blast and the Hundred go without potential player reward?”

You’re right about the IPL teams making mistakes by overspending on those ‘known’ players and I’ve already alluded to that in previous answers above too. I can’t go into too much detail about discussing how I’d go about recruiting under the radar players but it’s obvious it’s something I look to achieve.

An extremely vague answer is that I build models for each tournament I work in which has expected performance levels for each player. Basically the model doesn’t care about reputation, it would value the same player whether their name is player X or, to use one of the names you gave, ‘Marcus Stoinis’. Then it’s all about marrying up created metrics from the model in conjunction with actual metrics, which helps to work out which players should go into the jigsaw, e.g. working out which players should positively contribute to the skillsets which I think are vital for that particular tournament.

In the IPL, I think there’s a lack of good recruitment analysts, or a lack of them who have a decent say in recruitment at least. I had some talks with one IPL franchise over the last month but they eventually decided they wanted a local analyst, and I wasn’t even charging that much for a year 1 price.

I don’t see any team recruit outstandingly well. Mumbai are decent for sure, and RCB have made huge positive strides, but I’m uninspired at best about the rest. Unless they get good people to work with them, I don’t expect much from the new teams and I wouldn’t be shocked if they made some gigantic mistakes which could easily be relatively cheaply prevented by hiring a decent recruitment analyst.

Couple this with a culture where reputation counts for far more than it should and it’s really not a surprise that many IPL teams recruit badly. Throw in the recency bias that you refer to as well, and it’s easy to see how teams overpay for mediocrity. The Big Bash timing of course, plus the prevalence of Australian coaches, helps Australian players get contracts which, in my view, they shouldn’t get on merit. The Blast in England is tough because as there are 18 teams, much of it isn’t televised. It’s hard to get the attention of overseas teams when all they have to go on is scorecards and the odd match on TV. It shouldn’t be an insurmountable problem, as teams should be doing their due diligence on players, but in my experience, most teams don’t.

Zac: “There’s been lots of talk about Australian wicket keepers in the last week or so and lots of it has been focused on the fact that Tim Paine is viewed as the best wicket keeper although he isn’t the best batter. Would you say there is an easy way to measure how good a wicket keeper is because it always seems like it’s purely a visual thing with no stats behind it when someone talks about how good a keeper is (Ben Foakes for example). Could you use maybe the amount of drops, missed stumpings and byes conceded by a wicket keeper I think this would be a flawed metric though. Interested to hear any thoughts you have on a better way or measuring the quality of a keeper?”

I received a couple of questions along these lines. Fielding metrics are miles behind batting/bowling metrics and keepers are no different. Football has gone some way to working out xG conceded for keepers and I suppose that’s a start but cricket is more complicated - for example if one keeper is good enough to stand up to bowler X and another keeper isn’t, how do you measure that?

Quantifying drops etc is also tricky. Some people count the runs scored after the drop as a cost which is absurd. I think the best way to measure that would be to work out the difficulty of the chance and the expected runs that the batter would be expected to score after that event and combine the two - for example if similar catches are taken 80% of the time and the batter was expected to score 10 more runs after the event, then the drop would cost 8 runs (80% of 10).

Generally though in cricket I think fielding is overvalued. Most players are within +/- 1 run in a T20 match with very few players being more than +1 or worse than -1. Maybe it’s different in a Test match, particularly in the subcontinent where a good keeper could have more value in a spin-orientated match, but I’m unconvinced anyone out there has an accurate measure of the value of keepers.

Girish: “On the IPL front, with the exciting draft picks coming up I would like to hear your opinion on whom RCB pick as their captain given a choice that they are able to pick their first and second choice options? Also I feel that Kohli while being a marquee player for RCB, does not necessarily make it to every XI now that he is not captain. What do you feel about that?”

Back to the IPL - I don’t really care who is captain for most teams because I don’t think most captains add much value unless 1) they are a genius (e.g. Dhoni) or 2) they are very willing to listen to data and take that into account when implementing their plans (e.g. Morgan). However I do anticipate that some teams (and probably at least one of the new teams) will overpay for a player because they want to make that player their captain - again, file that under big mistakes that teams will make.

On your Kohli question, with Kohli no longer being captain, in my view it is extremely difficult to justify RCB or in fact any team paying him 10+ Crore. Anyone who disagrees should have a good look at his boundary percentage across the last three seasons, particularly against spin.

Also Girish: “My question for your revolves around the various formats that are floating around now. Test, One Day, T20 and now T10. Do you think Cricket can accomodate all these formats without it becoming an overkill? Are we now reaching , dare I say - Cricket saturation? Is the multi-format player a dead concept? Do we now need to spin off T10 and maybe even the 100 into a separate sport by itself?”

I’m a big believer in the diversification of formats because it’s virtually impossible to be elite at all three major formats. For example, very few high quality Test batters are great T20 hitters - they are mostly anchors, and that’s because the difference in skillsets required is so huge. You can’t expect someone to hit at 50SR in a Test match and then a week later in the IPL strike at 150+.

I don’t think we are reaching overkill or saturation, but I wouldn’t be surprised to see, for example, 50 over cricket take a back seat over the next 10-20 years. Many people now have lower time availability or attention spans so the shorter formats automatically become more popular to much of the younger generation of fans, at least.

Lindsay: “Warner was player of the tournament at the T20 World Cup Do you think a team will pick him up in the next IPL Auction?”

He’ll definitely get sold and probably for good money. Whether I would pick him up is a different debate - I’ve already written an article questioning whether he’s top level and of the 289 runs he scored in the World Cup, he hit 65(42) against Sri Lanka and 89(56) against West Indies which was basically a testimonial match. That’s over half his runs scored against teams with mediocre bowling attacks at international and franchise level, so I’m sceptical as to whether he’s back to his best, as the media narrative would like you to believe.

Chris: “Do you think T10 could be a good warmup for a T20 World Cup? Never thought about it until recently, but in a session you could have 2 games and a good way for players to hone skills. Death bowling and attacking batting”

Absolutely - to teach players the value of intent and the lower value than many perceive towards wicket preservation. I’m loving my time in the T10 League currently and the general increase in intent makes for a great watch, in my view.

Muhammad: “Can you review PSL 6, what went wrong for whom & when? PSL 7 Draft is lined up so suggest what could be done to yield better results?”

I’ll write an article on this in the not too distant future. I think the draft is supposed to be pretty soon so I’ll get something sorted in advance of that.