In the age of Peak TV, the quest for the instantly perfect can be the enemy of the eventually good.

Once upon a time, the decision to commit to a TV show had as much to do with what it might become as what it was at the moment. Rare is the show that's a Hall of Famer right out of the box, so there's always been guesswork involved: Do I like this premise? These characters? Do the parts I enjoy seem likely to stay while the parts I don't disappear? What's the best possible version of this show, and what are the chances it actually becomes that?

In those days, a patient viewer might have endured the early growing pains of shows like Seinfeld, Buffy, Parks and Recreation, or even Breaking Bad, and in time been rewarded by those shows transforming their potential energy into the actual, kinetic kind.

These days, though, with so many viewing choices out there — not only the best current shows just a click away, but many of the best shows of all time — the level of patience required to let shows figure themselves out seems in shorter and shorter supply. When I publish reviews of promising new shows that aren't quite there yet, the reaction is often, Oh, well. Who has the time if it's not already great?

I'm often as guilty of this as my readers. In less busy times, if a show demonstrated even the faintest spark of something more exciting, I'd give it at least a half-season, if not a whole one, before giving up, and even then would often come back a time or three in ensuing years, just to be sure it hadn't pulled a Star Trek: The Next Generation and required an especially long period of time to get its act together. Eventually, that grace period dwindled down to four or five episodes, and the last few years, I've found myself writing shows off as Not For Me after only the pilot, just because it's not possible to keep up with everything.

Usually, my instincts turn out to be right — I don't regret the hours I didn't spend watching more of Minority Report or Blood & Oil — but sometimes shows fall through the cracks. I forgot to DVR the second episode of Limitless, then decided it was more trouble than it was worth to keep up with a show whose ceiling at best seemed to be competent procedural with some sci-fi flavor. Based on the handful of episodes I watched this week (including the Ferris Bueller riff and the two-part season finale), I was wrong in my assumptions, because Limitless seems fun and creative enough to transcend its format. If CBS orders a second season, I'll be watching.

Which brings us to the show that's replacing Limitless on Tuesdays at 10 starting next week (and will also be airing episodes Mondays at 10 starting the week after that): the fifth and final season of Person of Interest, another series I dismissed too soon.

Back in the fall of 2011, I liked the idea behind PoI, in which reclusive tech billionaire Harold Finch (Michael Emerson) built an artificial intelligence program to monitor the entire country for terrorist activity, then set up a backdoor into the system which would send him Social Security numbers of people who might be either victims or perpetrators in small-scale crimes the government otherwise considered irrelevant. This was a timely twist (two years before Edward Snowden, even) on the kind of vigilante drama CBS had been making for decades, it seemed a good post-Lost role for Emerson, and on paper the cast around him was strong, including Jim Caviezel as Finch's strong right hand John Reese and Taraji P. Henson as NYPD vet Joss Carter, who didn't know quite what to make of this man in a well-tailored suit who kept shooting bad guys in the kneecap.

But I gave up on it after a half dozen episodes or so, out of frustration with the blandness of the weekly cases, the dumb idea of keeping Carter out of the loop of what Finch and Reese were really up to, and, mainly, the borderline-comatose nature of Caviezel's performance in the lead role. What was meant to be cool reserve in the manner of a Clint Eastwood instead came across as boredom with the material. Given the other early creative stumbles, I could understand that attitude (even if it wasn't Caviezel's intention), and if the show's hero didn't seem to care about any of it, why should I?

So I moved on, and when I occasionally heard TV critic friends talk about how much PoI had improved, I made a mental note to revisit it some day... and then just didn't. That's another factor of Peak TV fatigue: it's hard enough to justify binging a cable drama that's heavily serialized and has only 10-13 episodes a season, but when you're talking about a broadcast network show that does twice as many a year and follows the old X-Files model of doing a bunch of standalone episodes for every mythology-driven hour, it threatens to feel like homework. After a couple of years, I assumed I would just never get back to it, even as what little I knew of it from the outside — say, that the cast had expanded to include a pair of actors I like in Sarah Shahi and Amy Acker — seemed promising.

But on a slow day early last month, I saw IGN's Eric Goldman and Matt Fowler tweeting their excitement about the show's impending return, and while our tastes aren't identical, they overlap enough that I finally asked them if there was a way I could easily skip past the early stumbles and boring parts and just get to what was good. As it turned out, Fowler had already written an article detailing a road map through each PoI season — primarily the mythology episodes, but also character spotlights and Number of the Week stories with particularly strong execution — that allowed me to race through the first couple of years while watching the majority of the more serialized third and fourth seasons before next week's premiere.

Thanks to that fine piece of TV critic service journalism, I found myself squeezing in PoI episodes (the previous seasons are all on Netflix) whenever I was between tasks at work or home, and the further I traveled down the map, the harder it became to move away from the show and back to more current episodes of other series.

Caviezel's whisper-acting never entirely grew on me(*), but the show learned to be playful with it at times (the new season's second episode allows his co-stars to do impressions of him, and of each other), and when Reese needed to be more emotionally vulnerable, Caviezel pulled it off. More importantly, though, the show's team, and world, expanded. Carter came into the fold (she was later written out, but as a strong conclusion to one of the show's better story arcs, and things worked out just fine for the future star of Empire), as did her blunt partner Lionel Fusco (Kevin Chapman), and in time the group added Sameen Shaw (Shahi), who had previously worked for the government preventing terrorist crimes on the "relevant" list, and Root (Acker), a cyber-criminal who reformed thanks to her religious belief in the power of Finch's machine. (The team also got a dog early in season 2, which did as much to humanize Reese as his friendships with the others did.)

(*) Other than maybe my episodic reviews of The Wire, nothing I've ever written has had as long a tail, comments-wise, as that original review of this show. Month after month, and year after year, some PoI fan would somehow find that review and express their displeasure with my dislike of Caviezel's performance. (Though, on occasion, someone would appear out of the Google ether to agree with me.)

The origin, capabilities, and potential danger of the machine also began to take on greater importance, as the show revealed itself to be a cyberpunk thriller masquerading as another CBS crime drama. Recent seasons have focused on a rival A.I. called Samaritan, which doesn't have the restrictions Finch placed on his machine, and which has insinuated itself into all corners of government and society as a result. Though neither computer system is played by an actor, both have distinct, vivid personalities at this point, and one of the series' best episodes (last season's "If-Then-Else") was told largely from the point of view of Finch's machine. The show's concerns about privacy and government surveillance have only become more pertinent, and the constant threat posed by Samaritan and its human assets adds an extra layer of tension to the Number of the Week outings.

Even in the final season (I've seen the first four episodes), there are still episodes that drag: the third episode is a pretty generic Number story, not even boosted much by new flashbacks to Reese's time as a CIA assassin. But the stakes and the proximity to the finish line for the most part have given the show freedom to experiment even more, like the powerful fourth episode, which reveals what's been happening to Shaw since she was captured last season by Samaritan's forces. The PoI creative team long ago figured out how to juggle the show's most CBS-friendly elements (ordinary people in trouble, stories that offer closure in an hour) with the parts that make it special (sci-fi mythology mixed with prescient government paranoia), and to have the confidence that everything would work together more often than not.

Patience is hard with any show that's not quite clicking yet, and may never. It's especially hard in an age when there are so many choices that require no patience at all to get to the good stuff. For all I know, taking the long way around with Person of Interest by watching episodes weekly as the show decided what it wanted to be when it grew up, I might have grown even more frustrated with it, and been unable to fully appreciate its turn for the better. But based on my experience in slower times hope-watching a Buffy or a Parks until they achieved their full potential, I suspect the satisfaction I'd have felt witnessing the improvement in real time would have outweighed whatever annoyance I felt wading through the duller Number episodes to get there.

Either way, I'm glad I gave it another chance, and that I had help in cutting a daunting task down to a more manageable size. Person of Interest was easy to write off as Not For Me in its early days, but it's Sure As Hell For Me as it gets ready to end.

Alan Sepinwall may be reached at

Alan Sepinwall has been reviewing television since the mid-'90s, first for Tony Soprano's hometown paper, The Star-Ledger, and now for HitFix. His new book, "TV (The Book)" about the 100 greatest shows of all time, is available now. He can be reached at