Monday, April 5, 2010

My strategy: Roster moves (Part 6 of 7)

This is the sixth in a seven-part series documenting my fantasy baseball strategy. In part five we discussed some ideas for drafting players. Now we'll examine some basic principles I follow when considering trades and waiver wire moves.

On May 10, 2009, I looked at Joe Saunders' season stats and was very pleased with my 18th round pick: He was 5-1 with a 2.66 ERA. Better yet, he was fresh off a complete game shutout victory over Zack Greinke and the Royals. But my celebration didn't last long. It was time for me to quickly get Saunders off my team--even if it meant flat out dropping him.

So how could this be? Joe Saunders had gone 17-7 in 2008, and so far in 2009 he was 5-1 with a 2.66 ERA. Why would anyone be in a rush to trade him, let alone drop him? Well, the answer is quite simple: In my world, he was no longer a "championship piece."

To trade or not to trade, that is the question

In theory, if you can develop a good system and then find the right personnel to implement that system, you can have a good team. The only problem, of course, is that this is easier said than done. You're going to have players who slump and guys who get injured, so there will be many times during a 6-month fantasy baseball season when you need to make roster moves, either finding guys on the waiver wire or making trades.

And really, we can simplify this a bit further. Every roster move you make is actually a trade. Either you're trading players with Mr. Other Manager, or you're trading players with Mr. Waiver Wire. I prefer to trade with Mr. Waiver Wire because I can usually rip him off.

There are several solid strategies for approaching roster moves. And I imagine that, at some level, most managers consider value over replacement player (VORP) whether they realize it or not. The Chairman has also posted about VORP. Essentially, with roster moves, you're expecting to add more stats to your team than you lose, and you want to factor in position scarcity to be more efficient in where you gain those stats. (For example, it's silly to trade your best 2B for another guy's best 3B when you have no other good options at 2B and a nearly-as-good 3B was already sitting on the waiver wire.)

Over the years, I've tried to add another wrinkle to my thinking when it comes to roster moves. I'm not sure if it's revolutionary, unconventional, or just plain reckless, but some seasons it seems to work okay. I call it the "champioinship pieces" philosophy.

Championship pieces

Back in part two, I explained the framework in which I try to accumulate points. Here's an excerpt from that previous post to give you a window into my thinking:
Every player on my team needs to have a clear purpose for being there. Even players on the bench. Players who start for my team are expected to produce now. Players on my bench are expected to produce when a starter can't go. Period.
Every manager wants his starters to produce now. That's nothing new. But I tend to put more emphasis on balance, and this includes having the strongest bench that I possibly can. Aside from the fact that it can lessen the damage done by injuries, a strong bench also gives me more opportunities to play platoons. 

In the excerpt above I also mention that "every player on my team needs to have a clear purpose for being there." When things are going very well, most players on my team are contributing in a big way, and the rest are at least contributing in a small way. It's rare to have this happen, but when it does, that's when my roster is completely filled with championship pieces.

What exactly is a championship piece? Well, it's hard to explain. I'm reminded of Justice Potter Stewart's famous quote from a 1964 Supreme Court ruling, "I shall not today attempt further to define the kinds of material I understand to be embraced within that shorthand description ["hard-core pornography"]; and perhaps I could never succeed in intelligibly doing so. But I know it when I see it."

But unlike Justice Stewart, I will try to further define championship pieces. Here goes:
  1. I want my team to put up great numbers
  2. I want my team to be balanced
  3. Any player proving to be a key contributor to either (1) or (2) is, in my mind, a championshp piece
Another important thing to keep in mind regarding the championship pieces philosophy is that it gains steam as the season unfolds. After all, early on it's anyone's guess as to how players and teams will perform. And if June 1 rolls around with my team mired in 9th place, I'm not exactly going to be viewing my players as championship pieces. However, if June 1 rolls around and my team is in the top 3 battling for a title, then I'm definitely viewing each player with one question in mind: Is he a championship piece or not?

If my team is going well in May, let's say in 1st or 2nd place, I might look at my roster and identify 16 or 17 guys as championship pieces. My goal is then to see if I can get up to 18, then 19 and so on. The main idea here is to not go backward. If someone offers to trade me a perceived championship piece, but I would have to give up a championship piece in return, then what has my team gained?

Let's take last season for example. If I had had the first pick of the 2009 UPL Baseball Draft, I would have taken Albert Pujols. (I ended up with the 8th pick and took Ryan Howard.) However, after my team bubbled to the top of the standings and my 17th-round pick Pablo Sandoval showed signs of becoming a good MLB hitter, an interesting development occured. If I had been offered Pujols for Sandoval, I likely wouldn't have done the deal. Why? Two reasons. 1) Sandoval added tremendous balance to my team with position eligibility at C, 1B, and 3B. 2) Pujols would have been a bit redundant, since I already had Howard. So yes, Pujols would have increased my team's talent level, but he wouldn't have improved its balance.

My modus operandi

In the Urbana Premier League (UPL) where I play fantasy baseball, I make relatively few trades with other managers. Here's why:
  1. Greater margins to improve my team exist on the waiver wire (I can more easily "rip off" the waiver wire than another manager)
  2. I want to be careful that any trade I make will fit into my team's overall system
  3. In general, I don't want to directly participate in making my opponents' teams better
Let me elaborate on that last point with an analogy. If you're on fire in the UPL, and I have a glass of water, and you're looking at me for help, then I'm probably just going to drink that glass of water.

Oh, I'm sorry, is that your house on fire? Welcome to the UPL, my friend.

Back in part four I opined that saves are a more reliable bet than win-loss ratio. As such, this can explain why I tend to favor trading starters for closers. It also explains why last season I cheered when the Thugs traded away Brian Fuentes for Clayton Kershaw--and then I cursed when he traded Tommy Hunter and Phil Hughes for Andrew Bailey. (Side note: My feeling is that the Bailey deal helped the Thugs' point total in 2009, but the Kershaw deal not so much. From a keeper standpoint, it will be interesting to see how this pair of trades looks in 2010 and the years ahead.)

Here's another example. In 2007, I had a healthy Brandon Webb pitching at a high level. However, I had a full staff of quality SPs, and as such, I started to question whether Brandon Webb was a championship piece or a redundant part with excellent trade value to help me land a real championship piece. After doing my homework on Jonathan Papelbon (including his full injury history, how he was managing his health, and watching interviews of him in the media), the Jabrones and I agreed to a deal: Webb for Papelbon. The Jabrones got what they wanted: One of the best starting pitchers in baseball. But in my estimation, Papelbon with his saves and rate stats would balance out my team better and I could win a championship without Webb. Fortunately, that year I tied for first, and yes, Papelbon's saves were critical.

What I Reallly Thought About Joe Saunders


Earlier I alluded to my Joe Saunders dilemma. He was 5-1 with a 2.66 ERA on May 10, 2009, but I wanted him off my team--even if it meant dropping him. So how did I reach this point? Easy. Saunders has a paltry K/9 rate, and it's hard to compete in that category with him on your team. Plus his only redeeming quality is his W-L ratio, which I've already stated isn't a very reliable stat. And finally, once I had determined he was no longer a championship piece, I started looking for championship pieces elsewhere. Time was of the essence, and I didn't want to waste it with Saunders being a drag on my roster.

And there was something else at play here. In the days leading up to May 10, my OF Manny Ramirez was suspended 50 games, and my 3B Aramis Ramirez landed on the 60-day DL. I needed to add a bat to my lineup. So I worked out a deal for Mike Cameron, who at the time had put up better numbers than Manny Ramirez over the first 5 weeks of the season. Granted, I didn't expect Cameron to keep it up, but I did hope he would be good enough to be my 4th OF and work him into my lineup a few times a week. In that regard, I did hope he would become a championship piece, and I did believe he was better than anyone on the waiver wire. (Those of you with good memories will recall that shortly after this deal was made during the week of May 10, both Saunders and Cameron went into a tailspin, so nobody "won" this trade. Although, it did win an unofficial UPL award for "most meaningless trade.")

Okay, now we're getting down to the end of this series. Next time in part seven we'll talk about the end game, where sometimes you need to chuck everything you know and just do whatever it takes to win.

8 comments:

Chairman said...

Greg - regarding the Fuentes-Kershaw trade: I still finished 1st in SV last season :-) And more importantly, I finished well under in IP, even with the addition of Kershaw (who the Dodgers coddled in September, costing me 4 starts). Plus, I thought that I had enough youth to move for another closer :-)

With regard to the endgame, we've talked about it quite a bit over on the UPL Blog in hoops in previous seasons - it's definitely is playing a huge in role in hoops this year, which I'm going to be posting about soon. I think that the keeper dynamic dramatically changes things. Love to hear your take on it.

Chairman said...

In terms of championship pieces, we have touched on a couple thoughts over at the UPL Blog. Obviously, there are different ways to assemble a team, but the goal is the same - somehow finish with the most points. This requires that you first figure out how many points you'll need.

One article looks at one way to look at overall scoring.

After you figure out your total, it's helpful to know how the different stat categories will look. The top of this article shows a way to figure out what an "average player on ca championship team" will contribute (for hoops, anyway). And I give a similar treatment for baseball, as well.

This should give you a nice idea of what stats you need from your players, though the 6x6 format adds for complexity, since there are other ways to get to the magic number of total points (I'm guessing it'll be 125 points with the new 13 team format).

Greg McConnell said...

Well, I don't remember all the specific numbers involved in the Fuentes-Kershaw trade. However, I seem to recall that Kershaw was initially a disappointment for you, getting unlucky with his W-L ratio. Your team started to sag a bit with saves, but you were able to recover there with the Bailey trade.

From a talent perspective, yes, I'd rather have Kershaw and Bailey instead of Hughes/Hunter/Fuentes. But, for me, it wouldn't have fit my m.o. to pull the trigger on the Kershaw-Fuentes deal if I'm trying to win in 2009.

Pauly said...

Dude I was going with you the whole time until you said you wouldn't deal Sandoval for Pujols.

There is no plausible explanation for that one -- especially because you knew that Sandoval was losing the C eligibility. It doesnt matter if you have Lou Gehrig as your first baseman, you still add Pujols you nut! :)

Sorry, but I had to call you out on that one.

Greg McConnell said...

Heheh. I had a hunch that I'd take some heat for the Pujols/Sandoval comment. I guess that I felt there were a few months last year where I had a "perfect storm" on my team that would make it tough for me to deal away Sandoval for Pujols.

How about this... I will openly challenge C-Lauff right now to offer me Pujols for Sandoval. Let's see what happens. ;-)

Chairman said...

Pauly - I'm right there with you. Especially with the structure of this league, where we run 2 UTIL slots.

At this point, we can't pay any more attention to anything that this idiot's spouting out :-)

Greg McConnell said...

So this is what makes fantasy baseball fun. We all have these different strategies, and then we get to test and refine them year after year.

I'll try one more time with this Pujols/Sandoval thought from last season. Here were my starters at C,1B, and 3B:

C - Victor Martinez
1B - Ryan Howard
3B - Aramis Ramirez

In a perfect world, these guys would never get hurt and would never slump. But chances are that something "bad" would happen to at least one. With Pablo Sandoval, I could replace any ONE of them. Now, let's look at it this way, including Albert Pujols in the picture:

Scenario 1

My team
C - Victor Martinez
1B - Ryan Howard
3B - Aramis Ramirez
Util - Pablo Sandoval

Other team
Albert Pujols

In this scenario, there is a decent chance that something "bad" will happen to at least one of these players. I'm not sure which one, though. However, if something bad happens to only one of the above 5 players, my team is still going to move forward. Maybe not as strong as before, but at least a Plan B is in place. Now let's look at it if I traded Sandoval for Pujols:

Scenario 2

My team
C - Victor Martinez
1B - Ryan Howard
3B - Aramis Ramirez
Util - Albert Pujols

Other team
Pablo Sandoval

If something "bad" happens to V-Mart or A-Ram, there's no way for Pujols to fill in for them. So in Scenario 1, any of those 5 players could get hurt and I'd still have a Plan B. In Scenario 2, there were 2 cases where I'd likely be struggling to find a replacement--and probably one not as good as Sandoval.

Also, keep in my that for much of last year Sandoval was my best 3B (A-Ram was hurt), so that's another reason it would have been a tough decision for me to trade Sandoval for Pujols. Plus I'm only referring to the months of May-July. By August, it was pretty apparent that C-Lauff was probably my main competitor, at which point my taking Pujols from him would have been a good idea at almost any cost.

One last point. It's much easier to replace a Util spot than an individual position (C, 1B, 2B, etc.) This is because it's hard to know what will be available on the waiver wire. In years past, there usually is someone fairly good on the wire, but who can predict which position will be available? There might be three nice options available at 1B, but that doesn't help you if you need a 3B. (Side note: Since the UPL has expanded roster sizes by 2 and added another team, that means that the 50 best players available on the waiver wire at any given time last year will no longer be available. That will be a big difference.)

I realize that much of this argument will be determined by value over replacement player and what you could reasonably expect to get for Pujols in a trade (i.e., if you got him "cheap" for Sandoval, what could you turn around and trade Pujols for in return?)

Oh well. This is especially funny since a Pujols for Sandoval trade would likely have been auto-vetoed anyway, and if not, it would have definitely been voted down by the league.

Greg McConnell said...

I'm also reminded of the time I went into an Ohio State football forum and wrote a long, detailed explanation of my idea to use a 4-team playoff to decide the national championship. I had all these rules and exceptions and so forth. I thought it would be great. But the first commenter simply wrote, "Have another rum, dumb!"

After giving it some thought, I finally decided he was right. ;-)