The value of a specialist keeper vs a wicketkeeper batsman

[Cipher]

Recently I saw a comparison thread between Ian Healy & Andy Flower.
The overwhelming majority went for Flower as the better test cricketer.
As someone who appreciates the craft of the gloveman I decided to research how much it really matters &
why have selectors prioritised keepers who bat well rather than simply selecting on their keeping skill.
It's a long post so if you can't be bothered I'll put the tl;dr at the bottom.

WK Batsman vs Specialist Keeper (Runs contributed)
Lets go with the Ian Healy vs Andy Flower poll for example. Flower averages 24 more with the bat to Healy.
How many byes does Healy save over Flower each innings? Maybe 2/3? (see figure 2 for byes comparison between some modern wicketkeepers)
Flower is still ~20 runs ahead per innings.

WK Dismissal %
According to this source Tracking the misses: the average modern test keeper drops a catch 15% of the time.
The article also details the number of missed chances (incl. missed stumpings) by some modern keepers.
The best was Boucher at only 10%. Gilchrist 12%, Dhoni 18%, Kamran Akmal 20% & the worst was Mushfiqur Rahim at 32%.
The statistician also states that Boucher & Gilchrist were in the single digits earlier in their careers.
Chris Read & Peter Nevill were the lowest at 7% but didn't appear in the table (I'm assuming their sample size was too small).
It is interesting to note that Akmal was considered quite a poor keeper whilst Dhoni is considered one of India's best but is only 2% better according to this metric.

Additionally the author wrote "Miss rates for leading wicketkeepers off spinners average around 30%, for both catches and stumpings, but it is only 10% for catches off pace bowlers. It can certainly be argued that keeping to spinners is the true test of a keeper." So I'm going to make some assumptions from these numbers (and you are welcome to debate them).

Firstly, the difference an elite keeper will make in catching chances from pace bowling may be marginal. If 10% is the leading miss rate from guys like Boucher, an ATG keeper specialist is unlikely to improve that rate by a significant margin because even they will still miss chances (no better than a generous 5% miss rate).

In contrast, retired keepers state that keeping to a spinner is more difficult & the article's stats support this. This could partly contribute why keepers from Asia have worse missed chance percentages, due to keeping to spinners & spin friendly tracks more often.
Given that the miss rate is higher at 30% for spin, this could be where elite keeping has a higher potential difference to the standard keeper.
Unfortunately I have not found statistics which show the difference between keepers for spin bowling only.
But hypothetically if an elite keeper had a better rate of say 10% compared to standard keepers, this could have an impact, more so for teams playing with 2 spinners.

Dismissals per innings
Looking at Statguru the median Wicketkeepers dismissals per innings is 1.66 (min 30 games).
This is just to illustrate historically how often keepers are involved in dismissals, not how good they are. Here's a few examples:

Alan Knott 1.545
Ian Healy 1.763
Jeff Dujon 1.800
Mark Boucher 1.975
Adam Gilchrist 2.178
Quentin De Kock 2.367 (highest)

It seems that modern day keepers get more dismissal opportunities than older era keepers, so for simplicities sake lets say a modern keeper makes 4 dismissals a match. So over a 5 match series that's 20 dismissals.

Impact
So lets compare how keepers of different abilities would perform. How many chances would each keeper need to get 20 dismissals for the series?
Bad Keeper (~20% overall miss rate) - C= 20/0.80 = 25 chances
Standard Keeper (~15% overall miss rate) - C = 20/0.85 = 23.53 chances
Elite Keeper (~10% overall miss rate) - C = 20/0.90 = 22.22 chances
GOAT Keeper (~5% overall miss rate*) - C = 20/0.95 = 21.05 chances

So by these calculations a standard keeper is going to miss just 1 more chance in a series compared to an elite keeper.
If you can believe that a Knott/Healy GOAT level keeper was at a ~5% miss rate* (Chris Read was 7%) they're 1 chance better than the Elites.

What does this chance missed mean for the team? Well it depends on who the batsman was, the batting conditions & other factors.
But we cannot predict whose wicket this chance takes so it is best to mark it as the average test wicket. Which historically from all nations & eras is worth ~30 runs.

In the Healy vs Flower example lets class Flower as a Bad Keeper & Healy as a GOAT Keeper
In a 5 test series Healy needs 4 fewer chances to reach 20 dismissals. Which hypothetically saves 120 runs over the series.
But with Flower's 20+ runs higher batting average compared to Healy, he only needs to bat in 6 innings in the series to match the run difference.

Of course this analysis doesn't take into account intangibles like: an elite keeper being able to take incredible half chances or the fear factor a batsman has of coming out to play spin with an elite keeper lurking behind you. Or what having a brilliant keeper does for the confidence levels of the teams bowlers and vice versa.
There is variance in the importance of the keepers ability to a team too. If the keeper has to bat less often & is more heavily involved with team dismissals their primary skill becomes more important.

tl;dr Standard wicketkeeper batsman on average provide more net runs to their team compared to elite specialist keepers.

Happy to discuss if you agree/disagree!

 

[TheJediBrah]

I appreciate the effort to quantify this but as you noted yourselves it is very rough and doesn't take into account a lot of, arguable even most, of the value or detriment the varying quality of a keeper provides

IMO:
Flower v Healy, they are such different cricketers it's almost impossible to definitively choose 1 over the other in terms of value. The cheat answer is "why not both" because if you have both available, of course you are going to pick Flower, but why does he have to be keeper? Even Zimbabwe took Taibu the minute they had a better option available and kept Flower as a bat. If you have both, you would probably pick both.

But if forced to make a choice I couldn't fault anyone going either way

 

[LangleyburyCCPlayer]

The problem with using % of dropped catches as a metric is that it doesn’t distinguish how hard a chance is, or even what a chance is considering a keeper may or may not leave a chance to their slips

 

[capt_Luffy]

I appreciate the effort to quantify this but as you noted yourselves it is very rough and doesn't take into account a lot of, arguable even most, of the value or detriment the varying quality of a keeper provides

IMO:
Flower v Healy, they are such different cricketers it's almost impossible to definitively choose 1 over the other in terms of value. The cheat answer is "why not both" because if you have both available, of course you are going to pick Flower, but why does he have to be keeper? Even Zimbabwe took Taibu the minute they had a better option available and kept Flower as a bat. If you have both, you would probably pick both.

But if forced to make a choice I couldn't fault anyone going either way

Tbf Taibu would have played most matches for his batting alone.

 

[Coronis]

The problem with using % of dropped catches as a metric is that it doesn’t distinguish how hard a chance is, or even what a chance is considering a keeper may or may not leave a chance to their slips

Very true. An example (not keeper but slips) would be Smith. With his stats and if you only saw some highlights you might think he was the GOAT slip.

 

[ataraxia]

Very true. An example (not keeper but slips) would be Smith. With his stats and if you only saw some highlights you might think he was the GOAT slip.

Isn't that saying the opposite thing? The number of chances Smith has is abnormally high, so the percentage of dropped catches better reflects his ability than catches per match.

 

[Coronis]

Isn't that saying the opposite thing? The number of chances Smith has is abnormally high, so the percentage of dropped catches better reflects his ability than catches per match.

I’m not sure of his drop rate actually.

 

[ataraxia]

I’m not sure of his drop rate actually.

According to Charles Davis, he took 22 catches out of 33 chances in 2023.

 

[sledger]

The starting point for me is that whoever is picked as the keeper (whether they are nominally a specialist or otherwise) needs to average at least 35 odd. If they can't manage that they just aren't worth the candle for me personally, whatever other benefits they provide. There is some nuance to this (I mean obviously you can't pick someone who is objectively shithouse at keeping), but I'd have a fairly decent bat who is a somewhat mediocre/questionable keeper (e.g., Bairstow) over a specialist keeper who can't bat to save their life (e.g., Chris Read) any day.

 

[SillyCowCorner1]

Was the Sanga-Prassana combo the best in terms of benefiting the team as a whole?

An exercise where we can look past the numbers and delve into the importance of each?

 

[Coronis]

According to Charles Davis, he took 22 catches out of 33 chances in 2023.

I believe the quote is, 2/3 ain’t bad

 

[ataraxia]

I believe the quote is, 2/3 ain’t bad

(it is when it comes to keepers taking catches)

 

[LangleyburyCCPlayer]

Very true. An example (not keeper but slips) would be Smith. With his stats and if you only saw some highlights you might think he was the GOAT slip.

There is all sorts of data available in cricket, you have black and white data such as data about matchups and periods of a match (which is very objective), I know there is data for expected wickets, but I don’t think there is anything like that for fielding, I wonder if cricket did have that, and we had data behind how good someone is behind the stumps, would it lead to more teams picking a Foakes type wicket-keeper instead of a Bairstow or Smith type? Obviously if you have a Gilchrist, it barely matters how they are with the gloves, and even so, he wasn’t bad was he?

 

[Cipher]

I appreciate the effort to quantify this but as you noted yourselves it is very rough and doesn't take into account a lot of, arguable even most, of the value or detriment the varying quality of a keeper provides

IMO:
Flower v Healy, they are such different cricketers it's almost impossible to definitively choose 1 over the other in terms of value. The cheat answer is "why not both" because if you have both available, of course you are going to pick Flower, but why does he have to be keeper? Even Zimbabwe took Taibu the minute they had a better option available and kept Flower as a bat. If you have both, you would probably pick both.

But if forced to make a choice I couldn't fault anyone going either way

The thing is though, keepers were on average taking no more than 2 dismissals an innings in the 20th century.
Even for all the plaudits for guys like Knott & Healy how often & how much is the their keeping skill actually impacting a game compared to a standard keeper? Considering that Andy Flower had a higher D/I rate (1.572) than Alan Knott or that Jack Russell only had a higher D/I rate of 0.009 over Alec Stewart are these unaccounted qualities even that important?
I would be interested to hear what qualities an outstanding keeper brings that occurs at the same consistency/impact that a WK Batsman's runs would bring. Certainly you don't want a mug behind the stumps but is a big sacrifice in batting ability worth the trade off in elite keeping skill?

Of course, both great cricketers in their own right. I suspect you would pick Flower if your team's batting was weak or your bowlers were not reliant on wicketkeeper assisted dismissals. Vice-versa for Healy. For ATG sides I think you would rather have Knott over Pant as an example.

 

[Cipher]

The problem with using % of dropped catches as a metric is that it doesn’t distinguish how hard a chance is, or even what a chance is considering a keeper may or may not leave a chance to their slips

I see your point, wouldn't the percentage partially reflect a keepers general catching ability though?
With the miss rates the article stated, very good keepers did perform better & all keepers face hard chances.
You are right that it doesn't highlight which keepers could fly through the air and take a screamer compared to the one's that green light chances to slips. Given we don't have a metric, how often do you think elite keepers take brilliant catches that a standard one wouldn't?

 

[Cipher]

The starting point for me is that whoever is picked as the keeper (whether they are nominally a specialist or otherwise) needs to average at least 35 odd. If they can't manage that they just aren't worth the candle for me personally, whatever other benefits they provide. There is some nuance to this (I mean obviously you can't pick someone who is objectively shithouse at keeping), but I'd have a fairly decent bat who is a somewhat mediocre/questionable keeper (e.g., Bairstow) over a specialist keeper who can't bat to save their life (e.g., Chris Read) any day.

This is my thinking too as far as selection for a normal side goes. I think if the value of an average wicket is much higher against an ATG batting lineup maybe the specialist keeper gets a closer look. Although when Australia was using Matthew Wade as the keeper rather than Peter Nevill that made me question things, Wade grassed far too many chances for my liking.

 

[Cipher]

Was the Sanga-Prassana combo the best in terms of benefiting the team as a whole?

An exercise where we can look past the numbers and delve into the importance of each?

In cases where WK batsman give up the gloves & has a big increase in runs that's certainly a choice worth looking at.
Alec Stewart is another example of this.
Although for others giving away the gloves didn't change much for their batting e.g. Matthew Wade.
Interestingly AB de Villiers had an even better average while he was keeping!

 

[Cipher]

There is all sorts of data available in cricket, you have black and white data such as data about matchups and periods of a match (which is very objective), I know there is data for expected wickets, but I don’t think there is anything like that for fielding, I wonder if cricket did have that, and we had data behind how good someone is behind the stumps, would it lead to more teams picking a Foakes type wicket-keeper instead of a Bairstow or Smith type? Obviously if you have a Gilchrist, it barely matters how they are with the gloves, and even so, he wasn’t bad was he?

I have a suspicion there is some data on it but it isn't privy to the public. Opta Sports have people recording fielding actions.
Given that the English have been continually picking WK batters despite some criticisms there must be some reasoning behind it.
That's why Gilchrist is such a gem, for most other keepers there is a trade off between keeping skill & batting. Les Ames may have been another.

 

[LangleyburyCCPlayer]

I have a suspicion there is some data on it but it isn't privy to the public. Opta Sports have people recording fielding actions.
Given that the English have been continually picking WK batters despite some criticisms there must be some reasoning behind it.
That's why Gilchrist is such a gem, for most other keepers there is a trade off between keeping skill & batting. Les Ames may have been another.

With the money in cricket these days, especially in the IPL, you're probably right, although with the current England setup rowing back on data recently, I'm not sure that's the reason we're picking batters who can keep!

 

[Coronis]

The starting point for me is that whoever is picked as the keeper (whether they are nominally a specialist or otherwise) needs to average at least 35 odd. If they can't manage that they just aren't worth the candle for me personally, whatever other benefits they provide. There is some nuance to this (I mean obviously you can't pick someone who is objectively shithouse at keeping), but I'd have a fairly decent bat who is a somewhat mediocre/questionable keeper (e.g., Bairstow) over a specialist keeper who can't bat to save their life (e.g., Chris Read) any day.

What would be the sort of level to pick over Bairstow, curious?