Who will you intentionally overpay for on your keeper list?

[TopChuckie]

As long as your other keepers have enough profit built in to make up for the potential loss it isn't a terrible idea. You want to build as much profit as possible onto your roster. A $10 loss sorta makes you do that with one hand tied behind your back. And if the draft pool is thin it may be the worst time to try and find profit in the draft itself.

And just because you told me not to think about it - I might have a higher view of CC Sabathia's upside at $1 than you do.

Thanks everyone. I feel better about my mathematically unjustifiable opinion.

And I'm a Yankee fan, so I love CC, but I just think usable SP's are the easiest thing to find in dollar days.

[Judge Jude]

The logical fallacy that some have is that the competition is won by who has the most profit, not the best stats. being too fixated on that will lead you to refuse to "overpay" - and have too many units left at auction's end.

I like to say, don't be the craziest guy in your auction - but at least stay close enough that you can still see him up ahead

[Ken]

The logical fallacy that some have is that the competition is won by who has the most profit, not the best stats. being too fixated on that will lead you to refuse to "overpay" - and have too many units left at auction's end.

I like to say, don't be the craziest guy in your auction - but at least stay close enough that you can still see him up ahead

If profit is appropriately calculated relative to inflation, and your stats are appropriately forecasted then the most profit will lead to the best stats, at least in the general sense.

Now obviously you have to make sure you are appropriately allocating $s so that you can fill your positions and fill your categories without overfilling one area relative to another. But that's a different discussion topic.

[Ken]

Thanks everyone. I feel better about my mathematically unjustifiable opinion.

And I'm a Yankee fan, so I love CC, but I just think usable SP's are the easiest thing to find in dollar days.

Ultimately if it is correct then it will be mathematically justifiable. But how we get there might be different than you expect.

If CC produces $6 of value but costs $1, then the comparison point is what else you can get for $1. If your thought is that your competition will leave other $6 pitchers available to you for $1, then for your league CC is not worth any more than $1, since that is the cost to get those stats.

Also, depending on inflation, keeping a $33 Segura, who you determine to be worth $23, may still be the mathematically correct choice if inflation is high enough. The keep/don't keep decisions are based on "profit" in terms value relative to inflated cost.

[cavebird]

Ultimately if it is correct then it will be mathematically justifiable. But how we get there might be different than you expect.

If CC produces $6 of value but costs $1, then the comparison point is what else you can get for $1. If your thought is that your competition will leave other $6 pitchers available to you for $1, then for your league CC is not worth any more than $1, since that is the cost to get those stats.

Also, depending on inflation, keeping a $33 Segura, who you determine to be worth $23, may still be the mathematically correct choice if inflation is high enough. The keep/don't keep decisions are based on "profit" in terms value relative to inflated cost.

This is the key. It is easier to find $1 guys who earn $6 than anyone who earns in the 20's---the same underlying principle is what causes inflation. It is easy to get the most "profit" from your lineup by filling it with all guys who cost less than $5. Then you finish last because you left $180 on the table. The key is to maximize production. Profit, in terms of what helps you win, is not linear. A $30 guy who earns $35 is worth far more than a $1 guy who earns $6.

[Mithrandir]

This is the key. It is easier to find $1 guys who earn $6 than anyone who earns in the 20's---the same underlying principle is what causes inflation. It is easy to get the most "profit" from your lineup by filling it with all guys who cost less than $5. Then you finish last because you left $180 on the table. The key is to maximize production. Profit, in terms of what helps you win, is not linear. A $30 guy who earns $35 is worth far more than a $1 guy who earns $6.

Is a $30 player who earns $20 worth more than a $1 player who earns $6?

[Ken]

Is a $30 player who earns $20 worth more than a $1 player who earns $6?

It depends entirely on your league whether it is “worth it” to keep that player.

In the abstract, in a startup or redraft with 0 inflation and assuming all owners make the best choices available relative to your valuations then no, the $30 cost means you threw away $10 and the $1 player allowed you to pocket an extra $5.

In a league with 50% inflation you will need to pay $30 at auction to get the $20 player so you aren’t throwing $ away relative to what you can do at auction.

And if you have teams keeping/drafting players below replacement value or improperly valuing certain players then the $6 player is “worth” less in terms of auction dollars.

[cavebird]

Is a $30 player who earns $20 worth more than a $1 player who earns $6?

What Ken said with the further explanation that it depends on what you can do with the remaining draft dollars. The key is to maximize production, not profit. Usually, they are the same, but it depends on how much inflation there is a keeper league. If all there is left that you can do with the extra $29 you have after trading the $30/$20 guy for the $1/$6 is buy a player with $10 of production, you do better to keep the $30/$20 player.

[Ken]

The key is to maximize production, not profit. Usually, they are the same, but it depends on how much inflation there is a keeper league. If all there is left that you can do with the extra $29 you have after trading the $30/$20 guy for the $1/$6 is buy a player with $10 of production, you do better to keep the $30/$20 player.

Actually it's much easier than that. If you calculate your inflation correctly and apply it before evaluating profit, you'll get the answer that looks at both.

If there's truly only players available at $10 or less, inflation will be astronomical. And in that case when you apply that huge inflation rate you'll see that the higher valued player provides more relative profit based on the dollars available in the auction.

For example, lets say inflation was huge, as you say, 100%.

$23 player is really a $46 player in your league now. So $46-$33 = $13 profit relative to auction dollars.
$6 player is really a $12 player. But $12-$1 = $11.

[Ken]

There's also another way of doing this calculation which I used to use. It may make more sense to others.

Lets say you have your $1 contract for a player worth $6 (player A) and you are comparing to your $33 contract for a player worth $23 (player B).

Assume you are just comparing the $33. And assume inflation of 100% (which is likely rare, but we're talking about leagues where there's only $10 players available, which is extremely rare too)

So for $33 you can have $23 of value by keeping player B.
Or you can have your $6 from player A plus $32 of auction dollars of which you can rationally expect $16 of value. $6 + $16 = $22.

So you keep player B in this case.

Turns out it's basically the same (and simpler) to do the math by just multiplying the value by the inflation rate before your profit calculations. So I don't use the "auction dollars devalued" method anymore, but it should come to the same conclusions.

[cavebird]

Pretty much this is accurate. But it is easier to just let the program do the calculations, lol.

[Ken]

Pretty much this is accurate. But it is easier to just let the program do the calculations, lol.

What's interesting is that "the program" often has flaws or biases. I personally prefer to do the calculations myself.

I previously used "lastplayerpicked" as one of my input sources, but the developer who created it stopped putting out rankings. He published his code and I went through it this year and found several places for improvement. He made several assumptions that were just wrong. The end result didn't change much but it was enough to make me question the whole thing.

The reality is that there is not enough demand more a more perfect valuation system for one to get created. Most of the time we think it's "close enough". Which can actually be said to be true since we are always basing our valuations on projections which are imperfect themselves.

But I like to improve those margins when I can.

[cavebird]

I am not that hardcore, lol. Programs are close enough to show inflation and a ballpark figure is good enough for me.