Not small sample size.

Because it's not a sample.  A sample is random across the entire data set.

What it really is is a snapshot.  A picture cropped to a specific point in time.  Like a window on the stock market.  It has it's ups and downs, but expand the window out and you get true performance.

Randomness yes.  Snapshot yes.  Sample size.....not so much.

That said, there's really no other explanation for the home swoon other than randomness.  Which, by the way, is all this game is about.

Posted by silentpadna on 3/23/2011 12:16:00 PM (view original):
Not small sample size.

Because it's not a sample.  A sample is random across the entire data set.

What it really is is a snapshot.  A picture cropped to a specific point in time.  Like a window on the stock market.  It has it's ups and downs, but expand the window out and you get true performance.

Randomness yes.  Snapshot yes.  Sample size.....not so much.

Poor choice of words, sure... but "SSS" are buzz words, so they gets (over)used to apply to everything.

Point being: 34 games is not (necessarily)  representative of the team's quality or lack thereof.

Point conceded.  Every once in awhile when I'm bored, I notice that term and I....just....can't.....help....myself.

I'm not a statistician, but I've studied some of it, both for engineering and business.  I'm kind of stickler for proper terminology.  I am very ignorant in a lot of areas myself, but not so much in this one.  When entering discussions about anything, the terminology I use demonstrates many times whether my points are ever even listened to, whether I make good ones or not.

I hear you.  I haven't dealt much with stats in about 10 years, but I appreciate your annoyance with the misuse of terms in general, and the sloppy overuse of "SSS" in particular.  :)

Posted by maxdeardorff on 3/23/2011 12:15:00 PM (view original):
You're probably losing a lot of games because your pitching is not so good (most SPs should have vs RHP in the 70s or at the very least high 60s).  It's not terrible, but to make up for it, you should either have a ridiculous offense (yours is just good) or a smothering defense (most of your players have sub-par defensive #s, leading to errors and - plays, both of which cost a lot of runs.)  You could probably make .500 by improving the defense and 90 wins by adding a true SP1.

All true.. Like I said I don't think this is a .600 club....Last season's FINAL W-L total is about where I feel they s/b. Considering some guys will do better than they should and other will be worse etc..If all live up to potential etc..they could challenge for a WC spot in this league..  But this terrible home record again just doesn't make any season (over the last part of previous season and the present one)..  If a team is going to STINK like this is doing, you'd think it would be both at home and on the road... Yet they play about as I expect on the road yet at home they are HOPELESS.

So you think the things you mention would apply whether at home or away..cause they are some what 'universal'.  Yet........

I love threads like this:

A) My team is acting funny, what is wrong?
B) Answer is given.
C) No. That's not right.  What's the actual answer?

Both of us flip a coin 30 times.  I lands on heads 10 times, tails 20 times.  You land on heads 20 times, tails 10 times.  Your response isn't "What the hell! That makes no sense!!!"

Posted by iain on 3/23/2011 12:49:00 PM (view original):
I hear you.  I haven't dealt much with stats in about 10 years, but I appreciate your annoyance with the misuse of terms in general, and the sloppy overuse of "SSS" in particular.  :)

It's just catch phrases or buzz words.   A few years ago WifS really cracked down on the forums.   Many of us screamed "censorship" and "freedom of speech" because they're easily understood.   Neither really applied because this a private site where they set the rules.   You can accept them or STFU.  But "restricting our right to communicate in the manner we choose" just isn't very catchy.

Snapshot sounds like a photography term.   Small sample size says "It's 12 games out of 162.  Calm yourself."

The reason...I reject the SSS most often is fairly simple. Whether a HBD Team wins or losses a game is a lot more than a simple coin flip.  Yes luck is a definite factor but there also a fair amount of 'tangible" stuff too..such as ratings, managerial settings, stadium effects etc etc etc..  So it's far more than a coin flip involved

Over a 35 game span....on average:

Your team will send close to 1,400 batters to the plate (not a small sample when looking at a team as whole and it's ability to hit)

Your team will have over 1,000 plays to make in field (not a small sample when looking at team's overall defensive abilities)

Your pitching staff will face close to 1,400 batters (not a sample when looking at team's overall pitching abilities)

So when looking at a team as whole 35-40 span really isn't really a small sample of it overall abilities.  Of course luck and the opposition will play a factor in the final numbers so some variations can be expected from one 35-40 game grouping to another BUT is not really a small sample size.

Now it's within these tangible things that I'm looking for help with and I thank those who have given an honest opinion about the team as whole with regards to it ratings etc..  Thing is besides improving the team through getting better players, is there something else going on.  Should I get better team speed to make up for less doubles...etc 

iain, for some reason, work blocks that page. Which is strange, they block almost no websites.

The -1 for 2B is not the reason you aren't winning at home.  If you don't accept the answer as "randomness" then I'm not sure there's anything else to say.

Also: 1400 is a small number.  140,000 is still kind of a small number, but is getting there.

One could argue that a player's entire career is not a very statistically significant amount of data.

Posted by burnsy483 on 3/23/2011 3:57:00 PM (view original):
iain, for some reason, work blocks that page. Which is strange, they block almost no websites.

The -1 for 2B is not the reason you aren't winning at home.  If you don't accept the answer as "randomness" then I'm not sure there's anything else to say.

Probably because of the fun little applet (in Java) described at the bottom.... the text:

The Law of Large Numbers says that in repeated, independent trials with the same probability p of success in each trial, the chance that the percentage of successes differs from the probability p by more than a fixed positive amount, e > 0, converges to zero as the number of trials n goes to infinity, for every positive e. Note two things:

 

1. The difference between the number of successes and the number of trials times the chance of success in each trial (the expected number of successes) tends to grow as the number of trials increases. (In fact, this difference tends to grow like the square-root of the number of trials.)
2. Although the chance of a large difference between the percentage of successes and the chance of success gets smaller and smaller as n grows, nothing prevents the difference from being large in some sequences of trials. The assumption that this difference always tends to zero, as opposed to this difference having a large probability of being arbitrarily close to zero, is the difference between the Law of Large Numbers, which is a mathematical theorem, and the Empirical Law of Averages, which is an assumption about how the world works that lies at the base of the Frequency Theory of probability.

 

The distribution of the number of successes in n independent trials with probability p of success in each trial is Binomial, with parameters n and p.

The controls on this applet let you change the number of trials, the probability of success in each trial, and toggle between viewing either the difference between the number of successes and the expected number of successes, or the difference between the percentage of successes and the probability of success in each trial.

I'd propose that a single season is a small sample size for a player.   That's why I don't get worked up when my Silver Slugger LF suddenly hits .258 with a .412 slugging in a season.  I don't like it but I understand that 1 hit every 5 games is the difference between .250 and .300.  I know, over the course of time, that he'll hit .300 on his career. 

If you're willing to accept that, you should be willing to accept that 9-10 players on a team can have an "off" year at the same time.    While that's unlikely, it's also possible.   

If you're willing to accept that, you have to accept that 1400 AB is a small sample size.

If you don't like the answers you're being given, you should tell us the answer so we can repeat it and make you happy.