Do Bad Reviews Have More Impact Than Good Ones?

One of those bits of unexplained, unsubstantiated bits of Internet "wisdom" I have run across holds that where books are concerned it takes a dozen good reviews to cancel out the effect of one bad one.

How did they come to that conclusion? As I said, it was unexplained and unsubstantiated. Yet simple math makes it clear that on Amazon, following a one-star review, it takes at least ten five-star reviews to get a better than 4.5 star average (4.54), and twenty such reviews to bring one’s average closer to 5 stars than 4.5 (4.76).

Of course, star-based rating systems are an attempt to quantify something only awkwardly quantifiable, but I have found myself wondering if there is not in those numbers some reflection of the reality. At the very least it seems to me that, all other things being equal, positive reviews individually have less effect than bad ones, and I can think of at least three reasons why this is the case.

1. Our personal experience--in many cases, long and painful personal experience--teaches us that most of what is out there runs the gamut from mediocre to plain dreck. Reviewers, considering it their job to grade content from good to excellent, and it increasingly seems, from excellent to outstanding, claim the complete opposite about any one thing. Therefore the odds are that anything we are being told is so great is likely overpraised--with the result that, all other things being equal, a less than glowing review is the more likely to be accurate from a purely statistical standpoint than a breathlessly praise-filled one. This may be reinforced by another aspect of the situation, namely that reviewers often seem to be incentivized to produce positive reviews--but it is less common that they have an incentive to produce negative reviews, and it would be my guess that most are less alert to those occasions when that incentive exists. (Considering popular works, for example, I have often noticed that reviewers tend to go out of their way to say negative things about works which are in some way critical or satirical of the political status quo--with the critical opprobrium directed at the recent Don’t Look Up an obvious example.)

2. It is more often the case that the critical, denigrating review will be specific in its comment, and support its assertions. Putting it bluntly, when we are pleased by something--perhaps the more in as we pleased by it--we are not moved to examine it very closely. Our impulse is simply to enjoy it.

Our reaction is the opposite when we are displeased--and communicate our displeasure to others. And it helps that we are more likely to see a need to demonstrate the validity of our reaction to others when we are hostile.

3. It may be that bad reviews are more entertaining than good ones--even when they do not necessarily work well as reviews. A positive review which is well-written is apt to really be about the thing being reviewed, rather than about itself or its author, and we respond to it accordingly. When someone does a good job writing a positive review we may respect the quality of the writing and the intelligence of the assessment, but our principal interest is in the item which received the review. Even if they showed some wit the entertainment value of the review is likely to be slight, because it was about the entertainment value of something else.

By contrast many an author of a negative review makes the review about itself, and for that matter, themselves, rather than what they reviewed. They put on a show of tearing apart the work they so disliked (or at any rate, tearing apart something because they are likely to get far removed from the thing they are actually reviewing). They become an insult comic giving a performance. Of course, this is rather narcissistic and mean-spirited approach, but if twenty-first century pop culture teaches anything else (and I’m not sure that it does) it is that narcissism and meanness for meanness’ sake have colossal audiences. And a reviewer who has entertained their audience is likely to be heeded even when not really reviewing the thing at all. (Indeed, this may have much to do with how people are, on average, quicker to volunteer a negative comment than a positive one--with obvious implications for the more casual sort of reviewer who supplies so much of the customer reviews at sites like Amazon.

Up against all that it seems only too predictable that one would need legions of positive reviews to subdue a handful of negative ones. For those sending their work out into the world, especially when they do not have a colossal media machine behind them, this is not a happy thing--a single troll easily doing a lot of damage, often entirely undeserved damage to the already slight chances of their work's being seen, and simply for the pleasure they derive from having caused harm. Indeed, one may wonder if for many it is worth bothering to pursue reviews anyway. If we need twenty five-star reviews to wipe out a single one-star rating, then the odds are massively against even the best work getting a fair hearing--and it may be well worth reconsidering the conventional wisdom about what publicity will do for a great deal of work.