Significance of the 'second innings denial' effect.

[Engle]

Think of a scale.

You add a little weight on one side of the scale because of weak batting (which causes one to bowl LESS in later innings )

But then, you have to add a little weight on the other side of the scale because of weak bowling (which causes one to bowl MORE in earlier innings )

Thus, making the net effect insignificant

 

[Starfighter]

Think of a scale.

You add a little weight on one side of the scale because of weak batting (which causes one to bowl LESS in later innings )

But then, you have to add a little weight on the other side of the scale because of weak bowling (which causes one to bowl MORE in earlier innings )

Thus, making the net effect insignificant

That's more like it. Now, which of these effects do you think may be more significant in tests, specifically considering Hadlee and McGrath as their strike rates and averages are similar and their styles and methods are not too dissimilar?

Or propose this scenario:

Bowler	BpM	1st	2nd	3rd	4th
A	255	171	174	135	63
B	236	134	140	106	97

Who takes more wickets, A or B, if their strike rates are the same (say, 51).

 

[Mr Miyagi]

Think of a scale.

You add a little weight on one side of the scale because of weak batting (which causes one to bowl LESS in later innings )

But then, you have to add a little weight on the other side of the scale because of weak bowling (which causes one to bowl MORE in earlier innings )

Thus, making the net effect insignificant

Not true.

Hadlee bowls 29 overs in the first and second.

He bowls 22 in the 3rd and just 8 in the 4th.

It is 51 vs 37. Significant.


Streak 27 and 26 in the first and second, he bowls 16 in the 3rd and just 8 in the 4th.

That is 43 vs 34.

It is significant.

Very very significant.

The overs drop drastically. Far more drastically than for McGrath, Marshall and Murali who all bowl far more overs in the 4th innings.

All players suffer this limit. Just like all players suffer competition for wickets. The effect for each player depends on the team that they played in.

By using a scale you're trying to make these limits dependant and adjusting to each other. They're not. They're very much independent from good bowling to good batting (or in great teams, both).

/thread

 

[Mr Miyagi]

That's more like it. Now, which of these effects do you think may be more significant in tests, specifically considering Hadlee and McGrath as their strike rates and averages are similar and their styles and methods are not too dissimilar?

Or propose this scenario:

Bowler	BpM	1st	2nd	3rd	4th
A	255	171	174	135	63
B	236	134	140	106	97

Who takes more wickets, A or B, if their strike rates are the same (say, 51).

 

[Starfighter]

You know, last time I checked an emoji wasn't considered a rebuttal.

 

[Bolo]

I'm going to ignore Murali. Spinners will bowl a greater % of overs in the second innings, and it's probably best to compare to other spinners.

We've got two standout quicks from strong (batting and bowling) lineups here and two from weak (batting and bowling) lineups. The first group has a standard pattern (relatively standard amount of overs bowled per innings). In the second group the pattern of percentage of overs bowled per innings is common. The total number of overs bowled per match diverges. This suggests Hadlee benefited in wpm more from the weakness of his bowling lineup than he lost from the weakness of his batting lineup, while the opposite is true for Streak.

Can anyone think of any other standout quicks from weak batting lineups we can look at to see if they are more like Streak or Hadlee?

 

[Starfighter]

I'm going to ignore Murali. Spinners will bowl a greater % of overs in the second innings, and it's probably best to compare to other spinners.

We've got two standout quicks from strong (batting and bowling) lineups here and two from weak (batting and bowling) lineups. The first group has a standard pattern (relatively standard amount of overs bowled per innings). In the second group the pattern of percentage of overs bowled per innings is common. The total number of overs bowled per match diverges. This suggests Hadlee benefited in wpm more from the weakness of his bowling lineup than he lost from the weakness of his batting lineup, while the opposite is true for Streak.

Can anyone think of any other standout quicks from weak batting lineups we can look at to see if they are more like Streak or Hadlee?

Thank you, this is the sort of answer I'm looking for.

Small sample size, but Rodney Hogg maybe? I can't think of any examples of quicks who averaged in the 22-26 sort of range for a good period without at least a decent batting lineup. Even Streak doesn't meet that criterium.

 

[Mr Miyagi]

I'm going to ignore Murali. Spinners will bowl a greater % of overs in the second innings, and it's probably best to compare to other spinners.

Unfortunately, while this may be true for Kane Williamson and Joe Root bowling their offies, Murali bowls more in the first innings than he does in the 3rd or 4th.

Murali as a spinner even tops the first innings charts, despite not being a seamer (I am happy to say this is a physical limit and lack of competition for wickets combined).

Murali bowls more in the 4th innings than Hadlee, Streak McGrath or Marshall. This is explained with two reasons, he had runs to bowl at, and he didn't have as much competition for wickets.

Murali - had the best of both worlds compared to the rest for wpm. :P

Runs to bowl at, and stuff all else taking wickets. Unsurprisngly, he is the highest wpm.

McGrath and Marshall had competitoin for wickets, this hurt their wpm more than Murali.

Hadlee and Streak didn't have runs, this hurt their wpm more than Murali.

 

[Bolo]

Unfortunately, while this may be true for Kane Williamson and Joe Root bowling their offies, Murali bowls more in the first innings than he does in the 3rd or 4th.

This is meant in a relative sense, not an absolute number of overs. I'm sure Muralis percentage of total team overs bowled in the 4th is higher, so he looks more like McGrath and Marshall despite playing in a weaker batting lineup.

 

[Mr Miyagi]

This is meant in a relative sense, not an absolute number of overs. I'm sure Muralis percentage of total team overs bowled in the 4th is higher, so he looks more like McGrath and Marshall despite playing in a weaker batting lineup.

His batting line up with Sanga, Dilshan, Jayasuriya, Jayawardene, Attapatu, Tilikiratne, Samareewa and Aravavinda was very strong comparatively to Hadlee and Streak, probably rivals WI, and yes everyone will be weaker than McGrath. :P

MUrali = good batting. Not weak batting.

Hadlee and Streak = weak batting.

McGrath and Marshall = good batting. With McGrath in just a simply outstanding batting team from Border's reign to Punter's.

 

[Bolo]

Thank you, this is the sort of answer I'm looking for.

Small sample size, but Rodney Hogg maybe? I can't think of any examples of quicks who averaged in the 22-26 sort of range for a good period without at least a decent batting lineup. Even Streak doesn't meet that criterium.

While it would be better, I don't think it is necessary to look at only ATGs. The quality of the bowler relative to his own team mates is more important. If we have someone who is better than his own team mates, we simply look at the overs bowled per innings and ignore the wickets taken.

I'd look up Hogg, but I'm not sure where. Site suggestions to get this done?

His batting line up with Sanga, Dilshan, Jayasuriya, Jayawardene, Attapatu, Tilikiratne, Samareewa and Aravavinda was very strong comparatively to Hadlee and Streak, probably rivals WI, and yes everyone will be weaker than McGrath. :P

MUrali = good batting. Not weak batting.

Hadlee and Streak = weak batting.

McGrath and Marshall = good batting. With McGrath in just a simply outstanding batting team from Border's reign to Punter's.

Lanka's batting was pretty horrible at the start of Muralis career. And it was never good away from home. How strong the batting is at its worst is more important to this exercise than how strong it was at its best. Nevertheless, on balance it's probably in the middle- more reason to ignore Murali in this comparison.

 

[Borges]

Generally, third/fourth innings scores are lower as the pitch wears out, and I would think that bowling more towards the end of the game brings greater rewards to the bowler.
As far as wickets per innings goes, this is mitigated by the fact that bowlers (particularly quick bowlers) tend to bowl less in the third/fourth innings.

Here are some quick numbers: overs, average, strike rate (1st-2nd/3rd-4th innings)

McGrath 2847/2028 22/21 52/52
Hadlee 2468/1185 23/21 51/50

Marshall 1636/1294 23/19 49/44
Murali 4421/2919 24/21 58/51

There is no significant difference in the case of the first two; but the difference for both Marshall and Murali are quite significant.

 

[Mr Miyagi]

Hadlee is just 15 average in the 4th innings :P

 

[Starfighter]

While it would be better, I don't think it is necessary to look at only ATGs. The quality of the bowler relative to his own team mates is more important. If we have someone who is better than his own team mates, we simply look at the overs bowled per innings and ignore the wickets taken.

I'd look up Hogg, but I'm not sure where. Site suggestions to get this done?

I just used cricinfo to get balls and averages and a bit of maths.

What we haven't actually been doing checking the batting and bowling strength, we've more been assuming it. You can do that in statsguru too, though it's a bit of a faff, that's where I found out, surprisingly, that Hadlee's teammates were actually weaker bowlers than Murali's (score one for Vaas) and even more surprisingly, that Marshall's were weaker than McGrath's.
You get balls in career/matches just from their profile (if you're addressing the OP) and so on, or innings breakdowns on statsguru, and for team batting and bowling search for all the matches including the player, then take that player's stats out to calculate their teammates' average.

The thing about including weaker players is: does the relative strength of the effects change with bowling ability? The more you think the more factors there are. It's just Hadlee was obviously brilliant and ran through sides with a weak batting team while Streak was less likely to blow sides away even if he had six Bradmans to bat for him.

BpM and WpM vs teammates rather than averages is an interesting one too. After all the gap between Trueman and Statham's averages (21.57 vs 24.84) is much smaller than their strike rates (49.4 vs 63.7), so that might affect the competition factor.

 

[Mr Miyagi]

Streak drops drastically in the 2nd incl the 4th too :P

 

[Bolo]

Hogg- same number of overs per match as Streak. Distribution is more extreme. His figures support the idea that quality of your own batting lineup plays a significant role in your potential wpm as a bowler.

32%-10.7 overs bowled in the 1st innings
33%-11.8 2nd
23%- 8.3 3rd
7%- 2.6 4th

35.4 total

 

[TheJediBrah]

So what's the point of this argument now? Just trying to quantify this effect?

The reason it came up earlier was the implication that Hadlee was "unlucky" in terms of wpm because he played in a weaker team, and hence is wpm was lower than it deserved to be in terms of his contemporaries. Which is clearly false. By playing in the team he did Hadlee actually managed a higher wpm than his contemporaries as a result, regardless of any "second innings denial effect".

No one was ever trying to say that the "effect" didn't exist, it's just clearly not decisive, and is outweighed when it comes to wpm by other factors.

 

[Mr Miyagi]

Generally, third/fourth innings scores are lower as the pitch wears out, and I would think that bowling more towards the end of the game brings greater rewards to the bowler.
As far as wickets per innings goes, this is mitigated by the fact that bowlers (particularly quick bowlers) tend to bowl less in the third/fourth innings.

Here are some quick numbers: overs, average, strike rate (1st-2nd/3rd-4th innings)

McGrath 2847/2028 22/21 52/52
Hadlee 2468/1185 23/21 51/50

Marshall 1636/1294 23/19 49/44
Murali 4421/2919 24/21 58/51

There is no significant difference in the case of the first two; but the difference for both Marshall and Murali are quite significant.

Ave E/R S/R


McGrath:

1st match innings 22.92 2.58 53.1
2nd match innings 21.20 2.49 50.9
3rd match innings 22.57 2.52 53.6
4th match innings 19.49 2.34 49.8

18% of his overs are 4th innings. This is literally at SK Warne levels in the 4th.

Hadlee:

1st match innings 22.17 2.69 49.3
2nd match innings 23.98 2.61 55.0
3rd match innings 23.00 2.59 53.1
4th match innings 15.62 2.37 39.4

6% of his overs are 4th innings

Streak:

1st match innings 30.88 2.62 70.6
2nd match innings 31.78 2.76 69.1
3rd match innings 18.38 2.35 46.7 *(this is interesting)
4th match innings 25.32 3.19 47.5

9% of his overs are 4th innings

Marshall

1st matching innings 24.41 2.89 50.5
2nd match innings 19.97 2.53 47.2
3rd match innings 19.63 2.69 43.7
4th match innings 17.65 2.33 45.2

11% of his overs are 4th innings

Shane Warne

1st match innings 27.63 2.83 58.3
2nd match innings 28.38 2.66 63.8
3rd match innings 22.67 2.51 54.1
4th match innings 23.14 2.62 52.9

18% of his overs are 4th innings.

Murali:

1st match innings 26.47 2.64 60.0
2nd match innings 21.39 2.30 55.7
3rd match innings 21.11 2.44 51.7
4th match innings 21.01 2.52 50.0

12% of his overs are 4th innings

 

[NotMcKenzie]

Hogg- same number of overs per match as Streak. Distribution is more extreme. His figures support the idea that quality of your own batting lineup plays a significant role in your potential wpm as a bowler.

32%-10.7 overs bowled in the 1st innings
33%-11.8 2nd
23%- 8.3 3rd
7%- 2.6 4th

35.4 total

Problem there with Hogg is that he was taking the most wickets during the W.S.C. era when his fast-bowling partners would have been Dymock and Hurst, and not the arguably better Lawson and Alderman (and Lillee and Thomson, but I don't know how many matches they played together); during that period, it was Hogg who was the spare bowler going in-and-out of the side IIRC. To take 41 wickets in six tests including two ten-WPM hauls suggests that he wasn't deprived of opportunity during his period of greatest success.

Now, question is, do you consider the post-Greg Chappell retirement side stronger or weaker than the 1978–79 Ashes side batting-wise? Keep in mind also that there was two series against the West Indies in this period, which may skew things by playing a clearly much, much better team.

 

[Starfighter]

Problem there with Hogg is that he was taking the most wickets during the W.S.C. era when his fast-bowling partners would have been Dymock and Hurst, and not the arguably better Lawson and Alderman (and Lillee and Thomson, but I don't know how many matches they played together); during that period, it was Hogg who was the spare bowler going in-and-out of the side IIRC. To take 41 wickets in six tests including two ten-WPM hauls suggests that he wasn't deprived of opportunity during his period of greatest success.

Now, question is, do you consider the post-Greg Chappell retirement side stronger or weaker than the 1978–79 Ashes side batting-wise? Keep in mind also that there was two series against the West Indies in this period, which may skew things by playing a clearly much, much better team.

Also needs context in form of total balls per match across the eras. This is the most important thing.

During the WSC era Hogg played 15 tests as leading bowler and bowled an average of 246 balls per match in a team with a batting average of 24.39 (wow, things were that bad!)

Post WSC Hogg played 23 tests, generally as third bowler, and bowled an average of 179 balls per match in a team with a batting average of 29.29

So despite playing in a stronger batting team which should have given him more second (third or fourth match innings is irrelevant) innings opportunities the fact that he had more competition and was effectively demoted was clearly more important to Hogg's bowling opportunities.