Gilchrist v Dhoni

[Sir Alex]

You could, if you so wished, remove not-out scores under 35 completely (simply counting them as outs would be even more grossly unfair) but I suspect it would make little difference to anyone's average.

After removing matches in which both scored less than 35 runs and remained not out, the averages are

Dhoni - 48.71 @ 90
Gilchrist - 35.54 @ 96.57

Usually statisticians multiply avg by SR to get batting index. Here the batting indices are

Dhoni - 43.87
Gilchrist - 34.31

 

[ColdSnow]

I hope I can enter this discussion now. And I would like to keep it simple as well.

I am going purely by what I have seen of Gilchrist and Dhoni. Gilchrist has accomplished so much more than Dhoni having been on the international circuit for so much longer. And right now, because of that I will have Gilchrist above Dhoni.

But I feel Dhoni has a multi-dimensional game which is very valuable - he can bat quickly but can also play the anchor. He also gives the "extremely cool under pressure" aura - which has a calming effect on the team. And I somehow feel Dhoni plans his middle-order innings as per the requirements of the situation. Gilchrist, OTOH, could only bat one way, whatever the situation.

If Dhoni continues like this, I will have Dhoni over Gilchrist after he plays the same amount of matches.

 

[Uppercut]

After removing matches in which both scored less than 35 runs and remained not out, the averages are

Dhoni - 48.71 @ 90
Gilchrist - 35.54 @ 96.57

Usually statisticians multiply avg by SR to get batting index. Here the batting indices are

Dhoni - 43.87
Gilchrist - 34.31

I think that statistical measure is a little unfair on Dhoni, but nevertheless, he absolutely dominates Gilchrist statistically. If Ikki or anyone else wants to make a case for Gilchrist being better all they can do is look at his longevity or performances in big games (neither of which Dhoni has had much opportunity to match).

 

[Ikki]

That's an impressive stat, however after a b it of thought it is still unfair on Gilchrist. Many scores of Dhoni's may be above 35, yet may be middling scores that had he played on he could get out to had he been opening, for instance.

A run of 40*, 37*, 45 ends up with an average much higher than a run of 70, 80, 40. Or to put it another way, Gilchrist could have many lower scores - still higher than 35 - that may have given him a not-out had he batted lower and that would have helped his average, whereas as it is he opens and he plays on to score more runs and eventually gets out which will actually harm him in this comparison.

Does that make sense, or do I need sleep?

 

[Uppercut]

That's an impressive stat, however after a b it of thought it is still unfair on Gilchrist. Many scores of Dhoni's may be above 35, yet may be middling scores that had he played on he could get out to had he been opening, for instance.

A run of 40*, 37*, 45 ends up with an average much higher than a run of 70, 80, 40. Or to put it another way, Gilchrist could have many lower scores - still higher than 35 - that may have given him a not-out had he batted lower and that would have helped his average, whereas as it is he opens and he plays on to score more runs and eventually gets out which will actually harm him in this comparison.

Just say Dhoni finishes on 21*. How many runs more would you expect him to get? Well considering that for every time he gets out he scores 50 runs, isn't it reasonable to assume that he'd add another 50 runs to his total? Given that the hard part is invariably starting your innings, that may be a little harsh if anything.

Why don't you try and work out what Dhoni's average would be if every time he was not-out he scored 50.76 more runs instead? Not that you need to. The average function does it for you.

 

[Uppercut]

Here's a nice fact- MS Dhoni has been not-out at the end of the Indian innings 35 times. They've only ever lost once in a match where their innings was completed with Dhoni not out (32 wins and 2 no-results).

I don't understand for a second why the hell you're trying to penalise him for that.

 

[Sir Alex]

That's an impressive stat, however after a b it of thought it is still unfair on Gilchrist. Many scores of Dhoni's may be above 35, yet may be middling scores that had he played on he could get out to had he been opening, for instance.

A run of 40*, 37*, 45 ends up with an average much higher than a run of 70, 80, 40. Or to put it another way, Gilchrist could have many lower scores - still higher than 35 - that may have given him a not-out had he batted lower and that would have helped his average, whereas as it is he opens and he plays on to score more runs and eventually gets out which will actually harm him in this comparison.

Does that make sense, or do I need sleep?

I think you indeed need some sleep dear fella! lol

I took 35 as the criteria because that is Gilchrist's average. An average, inter alia, is a statistical measure of what a batsman "could" be expected to score in an innings right? So whatever Dhoni or Gilchrist scores above that should be considered and given justice in totality. In your example itself, had it been Gilchrist who was scoring those 70,80,40 then his average would be 63.33, which then I would have used to discount the earlier figures of 40*, 37*, 45 (like 35 used in Dhoni vs Gilchrist case above).

I agree with what Uppercut said. Adding to longetivity and World Cup wins, I would also suggest "An aura of destructive strokeplay". Dhoni indeed possessed that in the first two years of his career. But now he is regarded as more of an accumulator rather than a destructive batsman.

 

[Ikki]

Just say Dhoni finishes on 21*. How many runs more would you expect him to get? Well considering that for every time he gets out he scores 50 runs, isn't it reasonable to assume that he'd add another 50 runs to his total? Given that the hard part is invariably starting your innings, that may be a little harsh if anything.

I am not quite sure how reasonable it is when he has 33 50s in 132 innings.

Why don't you try and work out what Dhoni's average would be if every time he was not-out he scored 50.76 more runs instead? Not that you need to. The average function does it for you.

But that's the point, we don't know how much more he'd score. But still, 35* 35* and 50 gives you a much better average than had he gone onto score 70, 70 and 50.

Here's a nice fact- MS Dhoni has been not-out at the end of the Indian innings 35 times. They've only ever lost once in a match where their innings was completed with Dhoni not out (32 wins and 2 no-results).

I don't understand for a second why the hell you're trying to penalise him for that.

I don't think we should penalise him for it, but I think it's still a factor either way that would help his average and harm someone like Gilchrist who opens; would it not?

After removing matches in which both scored less than 35 runs and remained not out, the averages are

Dhoni - 48.71 @ 90
Gilchrist - 35.54 @ 96.57

Usually statisticians multiply avg by SR to get batting index. Here the batting indices are

Dhoni - 43.87
Gilchrist - 34.31

Wait, I just did these and I got difference results:

Dhoni: 4924 - 199 (the runs he scored when not out and less than <35)/ 132-(35-12 (number of innings not out and < 35)) = 43.4

Gilchrist: 9619 - 95 / 279 - (11-5) = 34.9

Index:

Dhoni: 39
Gilcrhist: 33.8

---

suffice to say I clearly underrated Dhoni, but I am not sure if the gripe I brought up is completely ignorable in the context of measuring not-outs accurately when we have players of different batting positions.

 

[Uppercut]

I am not quite sure how reasonable it is when he has 33 50s in 132 innings.

How many runs would you back him to score before getting out then? Statistically he generally scores 50 runs before getting out, so that's surely the number we should use?

 

[Sir Alex]

Here's a nice fact- MS Dhoni has been not-out at the end of the Indian innings 35 times. They've only ever lost once in a match where their innings was completed with Dhoni not out (32 wins and 2 no-results).

I don't understand for a second why the hell you're trying to penalise him for that.

That is an exceptional piece of statistics. Wow.

 

[Ikki]

How many runs would you back him to score before getting out then? Statistically he generally scores 50 runs before getting out, so that's surely the number we should use?

Yes, because statistically he has many scores well below 50 that don't count as a full innings because he didn't get out. He can get a series of those scores and it will give him a much larger average.

Whereas I think it is reasonable to assume even if he goes onto double the runs he makes in some of those not-out scenarios his average would actually be poorer just for having batted long enough to get out. I don't think it's very controversial to say that you are much more likely to get not-outs in the 30-40 range than keeping them for higher scores, right?

If Dhoni got:

35*, 35*, 35*, 50, 35*, 35*, 35* and 50 his average would be 155.

If he actually tripled his not-out scores

105, 105, 105, 50, 105, 105, 105 and 50 he would still have a lower average of 91.

That seems to be the only problem for this measure, for Gilchrist is rarely ever likely to finish a game not out between scores of 35-70 - in fact, of his not-out scores he only has 3 in that range, for example.

 

[Sir Alex]

Wait, I just did these and I got difference results:

Dhoni: 4924 - 199 (the runs he scored when not out and less than <35)/ 132-(35-12 (number of innings not out and < 35)) = 43.4

Gilchrist: 9619 - 95 / 279 - (11-5) = 34.9

Index:

Dhoni: 39
Gilcrhist: 33.8

---

suffice to say I clearly underrated Dhoni, but I am not sure if the gripe I brought up is completely ignorable in the context of measuring not-outs accurately when we have players of different batting positions.

Ikki, When you are excluding not outs, you got to reduce that from both innings played as well as not outs. Here the revised number of innings would be 120 (132-12) and the number of not outs would be 23 (35-12) So the revised number of runs 4725 (4924-199) should be divided by revised number of dismissed innings 97(120-23).

It is probably uncharitable to even consider a statistical comparison as the only barometre of judging two cricketers. There are infinite number of variables which cannot be quantified like weather, pitch, opposition, batting strength of own team etc. At the best these can be only an indicator in that analysis. However here the gap between Dhoni and Gilchrist is somewhat substantial enough to say Dhoni is statistically superior as a batsman than Gilchrist.

 

[Ikki]

Ikki, When you are excluding not outs, you got to reduce that from both innings played as well as not outs. Here the revised number of innings would be 120 (132-12) and the number of not outs would be 23 (35-12) So the revised number of runs 4725 (4924-199) should be divided by revised number of dismissed innings 97(120-23).

Ah, I see. Thanks.

 

[Uppercut]

Yes, because statistically he has many scores well below 50 that don't count as a full innings because he didn't get out. He can get a series of those scores and it will give him a much larger average.

Whereas I think it is reasonable to assume even if he goes onto double the runs he makes in some of those not-out scenarios his average would actually be poorer just for having batted long enough to get out. I don't think it's very controversial to say that you are much more likely to get not-outs in the 30-40 range than keeping them for higher scores, right?

If Dhoni got:

35*, 35*, 35*, 50, 35*, 35*, 35* and 50 his average would be 155.

If he actually tripled his not-out scores

105, 105, 105, 50, 105, 105, 105 and 50 he would still have a lower average of 91.

That seems to be the only problem for this measure, for Gilchrist is rarely ever likely to finish a game not out between scores of 35-70 - in fact, of his not-out scores he only has 3 in that range, for example.

There's good reason- if he'd had the chance to play on from 35*, how many more runs would he have scored? Considering that he generally scores 155 runs before getting out- and that when having to start his innings repeatedly- it's reasonable to assume he would have scored another 155, surely?

 

[Sir Alex]

That seems to be the only problem for this measure, for Gilchrist is rarely ever likely to finish a game not out between scores of 35-70 - in fact, of his not-out scores he only has 3 in that range, for example.

Ikki, I stated in my last post why I took 35. It is because that's Gilchrist's average.

Also note that despite having so many notouts Dhoni has played more balls per innings than Gilchrist.

 

[Ikki]

It is probably uncharitable to even consider a statistical comparison as the only barometre of judging two cricketers. There are infinite number of variables which cannot be quantified like weather, pitch, opposition, batting strength of own team etc. At the best these can be only an indicator in that analysis. However here the gap between Dhoni and Gilchrist is somewhat substantial enough to say Dhoni is statistically superior as a batsman than Gilchrist.

Notwithstanding my criticism, I agree. Once this kind of gap is established it is fairly accurate to make a generalisation. Using the formula in this thread Dhoni is on the heels of Tendulkar, even.

 

[Sir Alex]

Notwithstanding my criticism, I agree. Once this kind of gap is established it is fairly accurate to make a generalisation. Using the formula in this thread Dhoni is on the heels of Tendulkar, even.

Of course Dhoni seems to be on the course to become the greatest ODI batsman ever if he continues in the same vein. He might not end up with the same amount of runs as Tendulkar does simply because he bats down the order, but that should not discount anyway his performances. I do feel that in the future, Dhoni will push himself upto No.3.

 

[Ikki]

There's good reason- if he'd had the chance to play on from 35*, how many more runs would he have scored? Considering that he generally scores 155 runs before getting out- and that when having to start his innings repeatedly- it's reasonable to assume he would have scored another 155, surely?

No, I don't think it's reasonable at all when his highest score is only 50! I think that's a somewhat flawed extrapolation that ignores the realities of cricket. Scoring 155 runs in one inning (if we consider said batsman an opener for instance) and scoring a series of 35* are completely different in difficulty.

 

[Sir Alex]

I just now voted and I am surprised to see more than 75% of people have voted for Gilchrist! Surely a major chunk of them would have disregarded that this is an ODI-only analysis.

 

[zaremba]

There's good reason- if he'd had the chance to play on from 35*, how many more runs would he have scored? Considering that he generally scores 155 runs before getting out- and that when having to start his innings repeatedly- it's reasonable to assume he would have scored another 155, surely?

With you on this, Uppercut, as you know from previous exchanges. The thing that your/our theory overlooks, of course, is concentration and the ability to build and maintain a long innings. But that needs to be weighed against the point about having to start your innings repeatedly, which is a strong one, and one which none of the "not outs flatter your average" merchants has ever begun to tackle. I would be genuinely interested to hear someone do so.