Plateau Players and Incremental Players

Continuing the second part of my 47-part series based on sas4's response to my post about Stage 4, sas 4 wrote:

An owner must master the concept of a 'plateau' player - the must have guy needed to mount an assault on any category. It is a fatal mistake to have spent all that money on J. Reyes and finish last in steals. Conversely, without a Reyes, you have a hard time making any progress in speed.
To go along with a 'plateau' player there must be the 'incremental' player, the Kinsler (I am hoping) who can get 20 steals and move you up rapidly through the category while not hurting you in other offensive areas.

As a reminder, sas4 and I were discussing the merits of an 80%/20% hitting/pitching split, as opposed to the more traditional 67%/33% split. I'm not entirely sure what the "plateau"/"incremental" players have to do with this discussion, but I'll take a stab at it.

Here is a look at the major league leader in a category versus the winning total for the category in my American League Roto league and my former National League Roto league.

Table 1: Winning Category Total as % of Category, 2006

League	Category	Individual High Total	League High Total	% of League High Total
A.L.	HR	54	278	19.4%
A.L.	RBI	137	1,046	13.1%
A.L.	SB	58	145	40.0%
A.L.	Wins	19	102	18.6%
A.L.	Saves	47	81	58.0%
N.L.	HR	58	249	23.3%
N.L.	RBI	149	984	15.1%
N.L.	SB	64	172	37.2%
N.L.	Wins	16	97	16.5%
N.L.	Saves	46	66	69.7%

So the idea of "mounting an assault on a category" only applies to stolen bases and saves...unless Jack Chesbro is going to be re-animated from the dead and repeat his 1904, 41-win season.

I'll go back in these same leagues (1999 for my A.L.; 2004 for the N.L.) to see if this rule actually applies.

Table 2: % of Category Winner's SB from one player

League	Year	Leader's best SB Player	LeadingTeam SB Total	% of Leading Player Total
A.L.	1999	42	173	24.3%
A.L.	2ooo	31	131	23.7%
A.L.	2001	56*	197	28.4%
A.L.	2002	32	115	27.8%
A.L.	2003	55*	140	39.3%
A.L.	2004	28	138	20.3%
A.L.	2005	32/54**	126	25.4%/42.9%**
A.L.	2006	58*	145	40%
N.L.	2004	70*	173	40.5%
N.L.	2005	26***	160	16.3%
N.L.	2006	40***	172	23.3%

* - A.L. or N.L. leader
** - Tie in category
*** - N.L. leader traded or acquired midseason

Of course, this is extremely anecdotal; I'd need data from hundreds if not thousands of leagues and a supercomputer to crunch all of those numbers.

For the most part, though, I'm not sure sas4's theory holds up. And, before I produced this table, I thought it would.

Immediately, there are three reasons I can think of why owning a "plateau" player isn't necessarily required:

1. In the American League, since the leader sometimes has a low SB total (Brian Hunter's 44 in 1999, Alfonso Soriano's 41 in 2002), it's entirely possible to spread the wealth in the SB category and not acquire a component player. Compared to the N.L., there are some pretty low winning totals.
2. Trades change the auction landscape quickly. Steals (and saves) are categories that teams sometimes give up on even in non-dump trades as soon as they realize they either cannot compete or are in a position where trading a SB guy will only lose them a point. Additionally, if you own Reyes and Roberts, you're going to trade one, and you will not own all 60+ SB that plateau player provides.
3. Finally, steals are far more likely to come from cheap sources. Chris Duffy came out of nowhere to steal 26 bases last year. In the A.L., Nick Punto and Willie Bloomquist stole 10+ bases and were cheap. Even if you find a power guy like this who will hit 20 HR, as Table 1 shows, you're garnering a lower percentage of total HR necessary to win the category.

There's another factor when it comes to plateau guys in categories like SB/saves, and that is the disappearing act some of these guys do. Memories are short in Roto, but anyone who paid big bucks for Chuck Carr, Brian Hunter or Alex Sanchez years ago probably saw his season go up in smoke. A 30 HR/120 RBI hitter generally doesn't fall off the map entirely like a 40-50 SB guy does. True, an 18/80 line from a 30/120 projected hitter does sting. But at least you're getting something. When Alex Sanchez disappears, you've spent $20-25 for nothing.

I'll summarize with two points:

1. You can win SB with incremental players only. In fact, spreading the wealth in the category might be a good idea, given the risk involved in so many speed only players.
2. If you do want to break the bank for a SB guy, make sure he's a complete player. This is a good place to look at the player's OBP instead of his BA and his other assets. A guy like Luis Castillo might not do much besides run and hit for average, but his OBP is probably going to keep him in the line-up for a long time to come. Meanwhile, someone like Scott Podsednik is already at risk somewhat, due to a poor OBP and the possibility of a platoon.