This is worth a read

Also, bear in mind that some come through free agency (non-drafted), and lots of peeps stick around for a heckuva long time (How long as Robert Horry been around?! Dikembe Mutumbo?!). Granted, lots drop out after 2 or 3 years. But a lot of scrubs are picked up to replace them via NBDL, undrafted free agents, etc. etc.

I believe I read somewhere that the average life span of an NBA career was only like 3.5 years. It was something way lower than I expected…........I guess I have to go find it now. Sorry boss…...

CDR ftw!

That’s what everyone said after Roy’s rookie season as well, and I, for one, feel that he improved greatly.

I see no reason why Roy won’t improve even more now that defenses won’t be able to double him as often.

Nickname?.

is that you can look at the draft from a historical perspective and see patterns. While any number of players can be the “next MJ”, historically that has not been proven to be plausible. That draftee has to be a Starter, and the best player on the team. Just like there are 4 aces in a deck, there are only 5 roster slots per team that are designated as Starter (barring injuries). The drafted player has to be good enough to crack the starting lineup, and then be able to deal with the pressure and responsibility that comes from that position, perform at a high level, and hope his surrounding cast is good enough so that he isn’t just a great player on a bad team (which doesn’t count for much).

So you are right in saying that there are no odds. That’s how Paul Milsap or Josh Howard slide to the end of the draft. However, you can look at the history of the draft and see how many all-stars were drafted in, say, the 3-6 slots, and how many came from the 21-28 slots. You will find a large disparity. Drafting at a particular spot does not guarentee an all-star. But if you’ve done your homework in the draft, and have a feeling about a player, you may have to move to a particular slot to get the guy. What you’re saying is it doesn’t matter whata slot that guy goes in, whether it’s 5 or 13, as long as you get him. I think Kevin Pritchard would agree with you, and so would I.

I think you are right that people make mistakes in using data from the past to make predictions about the future… And, you are absolutely right that the fact that 50% of previous #10s make the rotation, does not mean that there is a 50% chance in this year’s class. I have, however, a couple of quibbles with your distinction between odds and statistics.

There are really a couple of things at play here. The first thing is that statistics is used to estimate probabilities. If we do not know the true probabilty of an event occuring, we can observe a number of experiments of that event and make an estimate. Usually, the larger our number of observations, the better the estimate; statistics has developed a methodology for quantifying our confidence in our estimate. Generally, the smaller the sample, the more uncertain we should be in our estimate, which is one reason to take the evidence cited above with a grain of salt. Of course, these methods require assumptions about the nature of the experiments. Typically, for example, we assume that each event is independent and identically distributed (i.i.d), which is unlikely to be the case for the experiment described in the article: given a player’s pick in the NBA draft, how often does the player make the starting five.

The second issue is that there are two notions of probability in the English language. One type of probability is the simple probability or “long-run frequency” of an event ocurring. For example, an introductory mathematics text would say, the probability that one draws an ace from a “well-shuffled” deck of 52 cards is 1/13. The second type of probability is known as subjective probability. A subjective probability describes an individual’s personal judgement about how likely a particular event is to occur. It is not based on any precise computation but is can be a reasonable assessment. In fact, much of financial markets and all sports betting is based on the interplay of subjective probabilities. Events that do not truly have long run frequencies. like the success of NBA draft picks, can still be assigned probabilities, but they are always subjective. Note that both of these types of probabilities can be converted into odds, which are just a different way to express the chance that an event occurs mathematically. For example, The odds of an Ace being drawn from a deck of 52 card is 12 to 1. Alternatively, according to Vegas, the odds that the Blazers will win the NBA championship next year is 25 to 1.

Getting back to the article, quantitative analyses, like the one above, can be used to enhance or improve the accuracy of the subjective probability that we assign to events. In fact, in most occupations make predictions is important and where the data exists now use some sort of statistical analysis to augment their judgement.