We’ve mostly talked about hitters so far, but 50% of our fantasy stats are obviously going to come from pitchers (at least in the vast majority of leagues). However, these pitchers who provide half our stats are only going to occupy about 30%-40% of our fantasy roster. This leads to an important question for formulating draft strategy: how do we value pitchers to maximize our rankings and get the most out of our drafts?Or in other words, when should we draft pitchers?Pitchers vs. HittersTo answer this question, there are a few concepts about pitchers that we must be familiar with. First, pitchers are generally more injury prone than hitters. Besides for the exceptions I mentioned in the last installment of this series (guys with injury histories, who play collision position, who get lots of HBPs, or who play with max effort), nothing is more likely to get a guy hurt than the unnatural act of pitching a baseball 90+ MPH. Even if you discount the fluky, catastrophic injuries that end seasons, many pitchers will miss a handful of starts or more with seemingly minor dings to various parts of their arm, or even their core or legs, all of which are vital to pitching at a major league level.Pitching stats are also much more team-dependent than hitting stats. Wins obviously require your team to score a certain amount of runs and, usually, your bullpen to be able to hold a lead. ERA (and to a lesser extent, WHIP) can often be affected by defense, bullpen stranding runners, bad timing, and many other things. Sure, R, RBI, and BA can be affected by similar things for hitters, but the variance isn’t as extreme and, since lineup quality is such a huge factor for R and RBI, it’s usually much more predictable.Between the injury risk and the team-dependent fantasy production, pitchers tend to have a wider range of possible outcomes and are therefore much more variable than hitters from year to year. Last year, in fact, 11 of the eventual top-25 pitchers in the Roto game came from outside the top 15 rounds of drafts, or weren’t drafted at all. At the same time, 10 of the top 25 starters drafted ended the year outside the top-35 at the position. This trend can be seen in historical data as well, which corroborates approximately a 40% “failure” rate among the first half or so of the pitchers taken, making them a very risky proposition, especially early in drafts.However, lest we think the pitching landscape is all risk from which no good can come, let’s take a look at the flip-side of the coin. 11 of the top-25 pitchers coming from outside the top 15 rounds of drafts means that quality fantasy pitching can almost always be acquired cheaply. Often times these pitchers won’t even be on owners’ radars pre-draft and can be had for free in-season off the waiver wire. This means we don’t have to waste many of our early round floor picks on pitching, as we can often get 1st or 2nd tier production later or even post draft.The cheap price of quality pitching, combined with its notorious variance, tends to also make pitching easier to trade for than hitting. We can often acquire one of those out-of-nowhere top-25 arms for very cheap if his owner thinks it’s a fluke that won’t continue. Conversely, if an established, earlier pick struggles early in the season, his owner may think he’s in that 40% failure group and let him go for pennies on the dollar (last year, this move would have worked well for a guy like Strasburg, for example, but not so much for Verlander). Meanwhile, elite hitters with safe floors always command top dollar in trades, so it is never a bad idea to err on the side of hitting knowing we can acquire quality pitching later in the draft, or mid-season via trade or the waiver wire.Additionally, I believe it is easier to make up ground in pitching categories than hitting categories in Roto leagues. A great pitching day (say 20IP, 2ER for example) can move our team ERA .05 - .1 in 1250 IP-max leagues (i.e. from 3.57-3.51), whereas a great hitting day (say 21/46) will only move our team BA .001 if we’re lucky. It’s simple math really, as it takes less to make an impact on 1250 IP than on the 6500 or so AB we’ll accumulate by the end of the season (depending on league setup)…If we really need to improve in K (which is really just K/9 in IP-capped leagues, more on that below) and WHIP, let’s say, we have the option of picking up non-closing RPs who will help without hurting us elsewhere, besides maybe Wins. However, if we need help in SB, most of our options to improve there will come with big-time downsides in HR, RBI, OPS and even BA. So once again, it’s smart to bank hitting stats early and know that you can make up pitching stats later.Finally, it’s important to realize just how few pitchers we actually need to provide the fantasy stats necessary to be in contention to win leagues. Assuming a standard setup of 1250 IP-max with 5 SP and 2 RP roster spots, we will come awfully close to reaching our IP max by simply filling those roster spots – with 5 solid SPs and 2 RPs. This is only 7 players out of our entire roster of 21+ players (again, depending on league), or usually less than 33% of our whole roster. Moreover, by actively playing the waiver wire throughout the season, we can get by with even fewer pitchers on our permanent roster. If we keep just 3-4 SPs on our roster we can use the last 1-2 SP spots to stream guys who are hot and/or have good matchups to help us fill up those 1250 IP. This isn’t a bad strategy to employ with our bench players anyway, as these streamers can become the undrafted guys who end up in the top-25, and earn a permanent spot in our fantasy team’s rotation.All that said, you may be thinking that I want nothing to do with any of the top-tier pitchers that require an early round pick. Quite the contrary! While W, ERA, and WHIP are team-dependent and hard to rely on year to year, there is one stat pitchers provide that is pretty bankable: Ks. In IP-capped leagues where everybody gets the same amount of IP to work with, the importance of K/9 (or K-rate) cannot be overstated. And it’s in the top tier of pitchers where we find the best and most reliable K-rates. Yes, we can make up for some of our lesser starters’ K-rates with elite-K RPs (which we’ll discuss next time when we look at position scarcity), but we want to have at least one SP who will provide that elite K-rate (9+ K/9) by himself. Because of this, I do think it’s smart to grab one of the top-tier pitchers early, but to then temper your desire to go too crazy with them after that in the middle rounds.However, the top tier of pitchers still cannot compete with the first two or so rounds worth of hitters in terms of floor, so I generally won’t consider taking my ace until the third round. This year, if all my offensive targets got sniped before my 2nd round pick, I would consider making an exception for Yu Darvish because his K-rate is so otherworldly compared to other SPs. But it would certainly not be my intent going into the draft, and I probably wouldn’t be too happy about it.I’m looking to get four bats and one arm with my first five picks, and ideally the first three of those will be safe, big impact bats. This year there’s a really strong contingent of potential aces available in the 3rd or 4th round, any of whom would make a great start to a fantasy rotation. Felix Hernandez, Cliff Lee, Max Scherzer, Jose Fernandez, Madison Bumgarner, Justin Verlander, and Chris Sale are all strong pieces to build a staff around, and there should be at least one of them available even if we want to use our first three picks on hitters. Let someone else take Kershaw or Yu or Wainwright, and watch them struggle to stay afloat in the hitting categories all year long.Once we’ve selected the ace for our fantasy staff, it’s time to go back to the hitters for a couple more rounds. Since so much quality pitching can be had later in the draft, or acquired cheaply (or freely) in season, there’s no reason to load up on multiple early round pitchers. I’ll basically skip over the next tier of pitchers completely, and then come back to them around the 7th or 8th round. At this point, there are still a number of guys with good K upside left that would make solid #2’s and 3’s for fantasy rotations. Again, we’re looking for K/9 since W, ERA, and WHIP are team-dependent and have a high variance year to year, so guys like Gio Gonzalez, Homer Bailey, Alex Cobb, Mike Minor, James Shields, Matt Moore, Julio Teheran, Michael Wacha, Danny Salazar, Shelby Miller, and Masahiro Tanaka should all be on our radar at this point. Ideally, we’ll be able to get two of these guys over the next few rounds and finish up our first 10 picks with seven hitters and three pitchers.Around this point you’ll want to start considering closers, but again, we’ll cover them and how they fit into draft strategy in the next installment. As for finishing out our starting rotation in the later rounds, we’ll want to grab another 2-3 starters between rounds 10-20. Even if you intend to aggressively stream pitchers through your last SP spot or two, you might as well fill those spots in the draft and give yourself as many chances as you can to get your hands on the numerous top-25 guys who come out of nowhere each year. These guys – basically any backups we take for our bench – are essentially lottery tickets: you hope they pay off but if they don’t you have no problem tossing them in the trash can. Because of this, I am fine with taking extra pitchers late in drafts (say after the 18th round) as long as you don’t have any offensive issues that need addressing. You only NEED pitchers to occupy about 1/3 of your roster, but there is no penalty for having more. Obviously, the more lottery tickets you have the better your chances that one will pay off.When searching for these later round pitchers in drafts, continue looking for that K-rate and for any plausible upside for a leap in value (young guy takes the next step, change in team situation, established vet finds his old form, etc.). Since a bad K-rate (think anything much below 7.0 K/9) can be such a category killer, and since so many of the established guys in the rds 10-20 range are there because of well-known mediocre K-rates, it is often better to take a chance on the unknown than to roll with the known and bad. Remember, if these guys don’t pan out they are easily replaceable. Some guys I’m looking for to fill out my rotation are (late) Justin Masterson, Jered Weaver, Andrew Cashner, Chris Archer, Ian Kennedy, Sonny Gray, AJ Burnett, Alex Wood, Brandon Beachy, Zack Wheeler and (later) Tyson Ross, Taijuan Walker, AJ Griffin, Dan Haren, Hector Santiago, and Josh Johnson.So when planning for your drafts, keep in mind that pitching is not only more variable from year to year, but also easier to acquire, easier to make up ground in, and you need less of it to provide the necessary stats than hitting. I generally like to take one ace early (rds 3-4), grab a couple #2’s somewhere between rounds 6-10, then snag another 2-3 SPs to fill out my rotation between rounds 10-20. In auctions, you usually want to spend about 1/3 of your money on pitchers, and I’m trying to more or less approximate that using draft rounds. When trying to decide which pitchers to take in those rounds, Ks (and particularly K/9) should be weighed much more heavily than W, ERA, and even WHIP, because the pitcher has much more control over them making them more predictable year to year. As always, the most important rule in draft strategy is to get the best value you can from each pick, so if a pitcher you like has fallen way past his ADP to the point that his value is too good to pass up, then I can endorse making exceptions to the blueprint presented herein. However, following these general guidelines should lead you to a team with a very strong hitting core, a couple of solid anchors for your fantasy staff, lots of upside from the bottom half of your staff, and enough excess hitting to trade for more pitching if none of that upside ends up panning out.