• Welcome to the Cricket Web forums, one of the biggest forums in the world dedicated to cricket.

    You are currently viewing our boards as a guest which gives you limited access to view most discussions and access our other features. By joining our free community you will have access to post topics, respond to polls, upload content and access many other special features. Registration is fast, simple and absolutely free so please, join the Cricket Web community today!

    If you have any problems with the registration process or your account login, please contact us.

Throwing away a start?

Bijed

International Regular
Having been involved in a bit of discussion recently about consistency vs conversion, the relative value of a 50 compare to a duck and a 100 compared to a 50, I got curious about which scores are made how often and all that, so I got a list of all the innings by top 6 batsmen (would have included nightwatchmen, but this should be a very small amount) played in the last 5 years (about 4,400 individual innings) and generated a graph showing the frequency with which each score from 0-319 has occurred. Given the commonly held perceptions about when a player is considered 'in', I was expecting to see maybe a steepish curve from 0 to the 30-40 range, which would be a local minimum, the rebounding back up a bit before dropping away as you get to the really big scores.

scoresdist.png

What you basically get is a sort of exponential curve down from the dreaded duck (which occurs massively more often than any other score) with a few sort of local peaks along the way - I'll see if those disappear if I include data from a few more years back.

Now, this wasn't exactly what I'd expected to see. Does this mean (and I ask this knowing that I can be quite bad at interpreting data) that the concept of getting in/throwing away a start isn't really a thing?
 

Daemon

Request Your Custom Title Now!
Not really, the decrease in scores are exponential, which shows that a greater percentage of players are getting from say 40 to 80 than 0 to 40. It shows that getting in is a thing.
 

cnerd123

likes this
Not really, the decrease in scores are exponential, which shows that a greater percentage of players are getting from say 40 to 80 than 0 to 40. It shows that getting in is a thing.
I think what Bijed means is that when we say 'throwing away a start', the inherent assumption is that a 'start' is meant to end up in a big innings, until and unless you 'throw it away'.

But the stats he has put out, it shows that very few players if any actually go anywhere with their starts. If a start was meant to end up as a big score, we'd have peaks at the low end (the starts that were thrown away) and peaks at the high end (the starts that were not). But we clearly don't see that, which probably means the whole 'throwing away a start' thing is just a fallacy.

It's kinda like the helpful conditions/minnow bashing fallacies, where it's just assumed that by appearing in helpful conditions/playing a minnow, the runs and wickets are granted to you and if you fail to take them it means you did something wrong. Kinda ignores that you still actually have to do it.

Just having a start doesn't mean you're guaranteed a big score 'unless you throw it away', according to these figures, it means that all you have is a start and you still need to work for all the other runs you wanna get.
 

TheJediBrah

Request Your Custom Title Now!
Not really, the decrease in scores are exponential, which shows that a greater percentage of players are getting from say 40 to 80 than 0 to 40. It shows that getting in is a thing.
wait isn't that the opposite of what it says? As in there's a greater percentage of players getting 0 to 40 than 40 to 80?
 

Starfighter

Cricket Web: All-Time Legend
I think 'throwing away a start' is a perception thing. If somebody falls for, say, less than twenty it will be though that the batsman's out of form or didn't play themselves in or it isn't their day or whatever, but once they've gotten a significant amount, say 35 or so, people are more likely to say the batsman 'threw it away' because 'clearly' if you've been in that long you should be capable of getting a bigger score. No surprise at all that the distribution of dismissals shows exponential decay.
 

Bijed

International Regular
Did no one get out for 6 in all of those innings?
It's actually 5 (sorry, the axes aren't that easy to read off precisely), but yeah, that was probably what surprised me the most. 100-ish people out for 4 or 6 each, but no-one at all on 5. Does seem weird.
 
Last edited:

Bijed

International Regular
Right, Daemon got me thinking that I was quite possibly looking at/for the wrong thing to draw any conclusions of whether or not getting a start is actually a thing, so I tried a different approach. Using the same criteria as before, but this time including all innings since 01/01/2001, I again generated the number of times each score has occurred. To smooth the data somewhat, I calculated a (forgive the probably incorrect term) five-point rolling total for each score, i.e. on the graphs below a score of 10 actually represents a score of 8-12 inclusive. For each of these, I calculated the number of innings that ended on a score greater than that range (so in the example given the number of innings that ended on 13 or more). I then divided this number by the first, then plotted the results of this against the scores that they related to. Basically, the x-axis is the score and the y-axis is how often the batsmen gets past that score once they're on it (or close to it, given the smoothing).

Kicking on.png

So the chance of a batsmen going on with an innings increases reasonably sharply and consistently up to the 30-40 mark, where it begins to fluctuate a bit, then dips quite sharply in the 65-75 range (a phenomenon known as the Root Effect), then actually goes right up until 100 is reached, when things go a bit crazy.
 
Last edited:

vcs

Request Your Custom Title Now!
Right, Daemon got me thinking that I was quite possibly looking at/for the wrong thing to draw any conclusions of whether or not getting a start is actually a thing, so I tried a different approach. Using the same criteria as before, but this time including all innings since 01/01/2001, I again generated the number of times each score has occurred. To smooth the data somewhat, I calculated a (forgive the probably incorrect term) five-point rolling total for each score, i.e. on the graphs below a score of 10 actually represents a score of 8-12 inclusive. For each of these, I calculated the number of innings that ended on a score greater than that range (so in the example given the number of innings that ended on 13 or more). I then divided this number by the first, then plotted the results of this against the scores that they related to. Basically, the x-axis is the score and the y-axis is how often the batsmen gets past that score once they're on it (or close to it, given the smoothing).

View attachment 23121

So the chance of a batsmen going on with an innings increases reasonably sharply and consistently up to the 30-40 mark, where it begins to fluctuate a bit, then dips quite sharply in the 65-75 range (a phenomenon known as the Root Effect), then actually goes right up until 100 is reached, when things go a bit crazy.
Nice. So people definitely play for landmarks (successfully, most of the time).
 

Adders

Cricketer Of The Year
Bijed, you know I love you man, but you're 21 years old.........go out and get drunk and laid ffs.
 

Top