Drafting SP vs. RP late

[UrbanatorNation]

I'm in a 12-team keeper league with 10 offensive spots, 7 P spots, and 7 bench spots. Over the past 3 seasons, its taken 1800+ Ks to be tops in that category in our league. We keep 4 hitters and 4 pitchers, and I'm keeping 4 SP:

Kershaw
Verlander
Lee
Kuroda

I typically like to draft 8-9 SPs, 2-3 CL and only a couple offensive bench players. Here's what my draft ususally looks like:

10 offense starters
2 offense bench
9 SP
3 CL
_____
24 total

I'm considering changing it up this year and instead of drafting those late SPs, drafting 2-3 middle relievers instead. Here's an example of what it might look like (conservative estimates):

3 late SP: 450 K, 24 wins, 4.00+ ERA
3 late RP: 200 K, 12 wins, 2.50 ERA (and maybe a few cheap saves)

The approximate drop in Ks, over the course of a season, would drop me 4-6 spots in that category. No use in chasing wins, although there would be a drop there too.

So is the nice ERA I'd get from relievers (vs. a SP blowing up for 5-6 ER) worth the loss of Ks and wins? Am I thinking about this correctly, or is there something I'm missing?

Thanks, WHIR

[Kimbos Beard]

The problem with projecting late RP's and SP's is that it doesn't take into account streaming your SP.

If you take a couple of good 8th inning guys like a Sean Marshall, my guess is that you are basically going to set it and forget it.

When you have late SP's you are generally more fluid with the roster spot. You can pick up guys when they are hot or when they have some favorable matchups coming up. You can't do that with the RP's you carry, you are basically married to them.

[UrbanatorNation]

I generally don't stream SP. I'll ride my lower SPs until its clear they aren't going to cut it, and then find new guys to fill in. I look at alot of different metrics, etc to try and make sure those late SP aren't total crap, and could maybe be worthwhile the entire season. However, that doesn't always mean they're good.

Maybe that 4+ ERA of those SP would be lower if I learned to stream more efficiently?

[Fenris-77]

You should definitely consider streaming at least one starter slot. i tend to polish off my pitching staff with middle relief guys too, mostly for ratio help, but getting key starts from streamers facing weak teams in tough ballparks is pretty useful.