Projection Theory Poll #2

[Guest]

Assume nothing else is known about the player, In the last three years in fulltime PA, he's hit 30, 20, and 10 HR. How many do you project him to hit this year?

this is just for discussion purposes. Please feel free to comment afterwards.

[msugray]

I find it psychologically interesting that at the moment I'm posting this, the votes are 50/50 on Poll 1 (hitter going 10, 20, 30) and 100% on Poll 2 (hitter going 30, 20, 10).

People want to regress on the high side and not account for a snap back on the low side.

By mere suspicion people assume that player in Poll 1 is a young player progressing and player in Poll 2 is a washed up veteran.

[cwk1963]

20. No significant injury time so I assume a veteran player who is regressing but has demonstrated a level of performance so a bit of a bounce back is likely.

[Shyguy30]

I voted for 20 but I'd be more likely to say 15. I like to use 3-year averages tempered by trends. If PT was the same, it's more likely this player hits 15 home runs than 10 or 20, at least in my mind.

[AllstonRockCity]

I find it psychologically interesting that at the moment I'm posting this, the votes are 50/50 on Poll 1 (hitter going 10, 20, 30) and 100% on Poll 2 (hitter going 30, 20, 10).

I actually missed the "fulltime P.A." caveat on the other post, but I think the overall point of what I was trying to say is still valid. And that is: there is data beyond just the actual stats that plays an important part in projecting a player. An old player declining and a yongster 'figuring it out', etc. would (i assume) certainly change how one interprets particular statistical trends.

[viper]

My true thoughts would be 15 but that wasn't an option. The first year is beginning to smell like an outlier but the last seems lowish. My thoughts are that something has changed.

[rotodog]

I find it psychologically interesting that at the moment I'm posting this, the votes are 50/50 on Poll 1 (hitter going 10, 20, 30) and 100% on Poll 2 (hitter going 30, 20, 10).

People want to regress on the high side and not account for a snap back on the low side.

By mere suspicion people assume that player in Poll 1 is a young player progressing and player in Poll 2 is a washed up veteran.

I actually just said something similar in the other thread before I read this, but a nice observation MSUGRAY...

[aburt19]

None of the above. I would normally go in the 14-16 range based on a partial recovery. I would hope to get him for the
price of 10 HR and think t hat I might have a profit.

[Mojo Jojo]

i was looking for the '8 hr' selection.

[Guest]

So it looks like here people liked a regression somewhere between prior years skill and the prior year/three year average.

The good news here - all other things being equal this is exactly how we calculate it.