Up For Debate – Are games too long?


I’ve long been standing on my soapbox and yelling to anyone who’ll listen that games have a length problem. Like the uninvited guest who starts taking their shoes off, they have a habit of outstaying their welcome. 

There’s a certain audience out there who’ll obsessively tot up the number of hours they’ll play a game and then attempt to use it as some sort of metric for value. As if value can ever possibly be attributed to length rather than quality. Would you rather chow down on a 60-foot dog turd or a 6-inch steak? I rest my case, you honour. 

This virulent line of thinking tends to emerge as $1 per hour played, meaning a brand new $60 game (or equivalent price in your territory) is only worth buying if it has at least 60 hours of gameplay. I get it, I do, but it’s a flawed line of thinking which fails to identify the value of a person’s time. There are very few games in the world which can offer 60 hours of truly fantastic content. Nearly all of them take a dip at some point, particularly in terms of the diminishing returns from new features, or mechanics which can genuinely change moment-to-moment gameplay.

I’ve circled back around to this theory again having just finished up the excellent Frostpunk. It is, in my opinion, one of the very few games with a near-perfect design; a tight structure which allows it to consistently build toward a crescendo. And then just ends. Frostpunk ends at that exact moment you’re having the best time. You want more of it, right there and then, but that’s your lot. 

That right there is how all single-player games should aim to be structured. There’s no time for familiarity to breed contempt. There are no rehashed ideas or filler content. It just gets better and better and bam, the credits roll.

Psychologists know all this though. So-called Peak End design has been the subject of much analysis over the past three decades. It all stems from a psychological study published by scientists Daniel Kahneman and Barbara Frederickson in 1993. Their particularly experiment honed in on the opposite of enjoyment though – it focused on pain. 

Test patients were asked to submerge their hands in tubs of water at various temperatures. They were then asked to put their hand in another tub, and then another. They were then asked to retrospectively evaluate the level of pain they felt. Kahneman and Frederickson discovered that patients’ recall of the events was wholly inaccurate. 

Patients strongly remembered the peak intensity of the pain along with the final three minutes of the test. Most patients did not recall the cumulative amount of pain nor the length of the test. 

As for what this means, take a look at the chart below. Both Patient A and Patient B were subjected to the same peak pain intensity. Patient A’s procedure stopped immediately after this point. Patient B’s procedure continued for 20 or so more minutes while the pain threshold was lowered. Patient B suffered the same peak pain and the most cumulative pain and yet, due to our tendency to remember peaks and ends, Patient B remembered the experiment as being less painful than Patient A.

So this led to a psychological phenomenon known as the Peak-End Rule, while all the bits we forget are known as ‘Duration Neglect’. 

Now if you’re still clinging on here, this entire philosophy can be flipped on its head and used to measure pleasure rather than pain. With entertainment, our opinion of something is heavily affected by its peak and by its ending. We remember snapshots of these moments and we kind of just discard the rest. But, a poor ending will lower our perception of the peak. 

Which leads us to Peak-End design. Big emotional highs can help offset low points. It’s a similar design philosophy which contributes to your standard feelgood movie formula. Our protagonist rises to a big high, has a tragic fall from grace, before rising to a crescendo at the end. 

Apply this to games and the ultimate goal is simple – a game peaking at the very end will leave the greatest lasting impression on us. I think Frostpunk achieves exactly this, to sublime effect. New challenges, rules and unexpected wrinkles are implemented at every time, doing just enough to paper the cracks which can be felt with the deaths of inhabitants or failure of plans. It builds to a crescendo at the tail end of the game, ending with a bang at the precise moment that each and every gameplay component has come together.

Another example with a less formulaic curve upwards is Sekiro: Shadows Die Twice. It is a game which is the embodiment of peaks and troughs for the players. At times you’ll feel frustrated and awful but, hopefully, the highs are good enough to keep you push boss after boss. It’s easy to envision an undulating line upwards for Sekiro, culminating in the final huge peak which most players will no doubt feel when finally toppling that rock-hard final boss.

That right there is Peak-End Design in action, a structure in place which tops itself at the final hurdle. The problem with Sekiro is it also comes with intense lows which can cause people to drop off entirely; it’s about ensuring the peaks outweigh the troughs.