There are two factors when it comes to a pitching value: how well a pitcher pitches, and how much a pitcher pitches. Pitcher A can dominate his opponents, but if he only lasts five starts before incurring a season-ending injury, his overall value for the season is limited. Similarly, Pitcher B can toss 250 innings, but if he's knocked around all season long, he won't be of much value (although, frankly, poor pitchers rarely throw that much).
Even though both factors are of paramount importance when it comes to assessing pitching value, I'd argue that peak tends to be overvalued in comparison to longevity. I'd expect that most people take the guy with the lower ERA over the guy with the higher inning count, even in cases where the higher inning count outweighs the difference in ERA. I'm not exactly sure how to go about proving this, though I'd expect end-of-season voting for the Cy Young Award reflects this to some extent, particularly when it comes to the starter/reliever divide.
It is for this reason, in part, that I believe Livan Hernandez to be undervalued. Hernandez, over his career, has maintained a 96 ERA+. He's been remarkably average -- slightly below, even -- over his 17-year career. However, he's also been remarkably durable. Over the 14-year stretch from 1998 to 2011, Hernandez averaged 33 starts a season. He led the league in innings pitched three years in a row, from 2003 to 2005, tossing 735 innings over that span. For his career, Hernandez has been worth 23.7 rWAR. That's nothing special -- that's what a consistently average pitcher would do over a 12-year career -- but it puts him in company with guys like Dave Stewart (also 23.7 rWAR) and Mike Flanagan (23.9 rWAR).
Hernandez is greatly undervalued for a couple other reasons though: his hitting and his defense.
Hernandez has spent the vast majority of his career in the National League, which is convenient, because he's an excellent hitter. Well, at least by pitcher standards. To date, he's come to the plate 1108 times, posting a .222/.231/.296 line (.231 wOBA, 32 wRC+). That's also nothing special -- there's not a convincing argument there that he should have converted to a full-time hitter or anything -- but it's markedly better than what the average pitcher can do with the bat. Considering that pitchers have hovered around a -10 wRC+ during Hernandez's years as a full-time starter, a 32 wRC+ from the nine-spot in the lineup is certainly pretty valuable. Indeed, Fangraphs has his work at the plate valued at 9.4 wins above replacement in total. If you add that to their valuation of his pitching (36.2 fWAR), that increases his total output by roughly 25%.
Now we come to the third aspect: pitcher defense. For this, we can turn to Defensive Runs Saved, of which there are three components: Stolen Base Runs Saved (rSB), Plus/Minus Runs Saved (rPM), and Good Fielding Plays (rGFP). The last component is negligible with respect to Hernandez, so let's focus on the other two.
rSB is calculated by accounting for stolen base success rates and frequency, with the assumption that 65% of a defense's control of the running game is in the pitcher's hands. Since 2003, Hernandez has been saved +9 runs with his control of the running game -- or roughly one win.
By rPM, the amount of plays a pitcher makes above or below what the average pitcher does, Hernandez has saved +24 runs since 2003. Considering that his best two seasons came in 2003 (+8) and 2004 (+7), the earliest seasons for which there exist such calculations, I think it's fair to assume that Hernandez rated pretty well by this measurement in the years that came before this data was available (1998 to 2002, in particular).
Even without that data though, Hernandez is at +34 defensive runs saved for his career. That, in addition to the 9.4 batting WAR, accounts for roughly 13 wins of added value. Considering this boost, and the fact that Hernandez's contributions with his arm alone are probably undervalued, I posit that Hernandez is the most underrated pitcher in recent history. He's definitely in the conversation, at least.