decision making biases — 1 thorough investigation

decision-making biases are, supposedly, the reason regular people choose poorly across all areas of their lives. i, by my own rough estimate, have made roughly forty thousand small decisions this year alone. carla has made detailed notes on six of them, in writing. the venn diagram of overlap, when i drew it on a napkin, was mostly one circle.

desk on the third floor, screen tilted away, mug at half. carla is upstairs in the q3 review with two binders and a face that has decided things before sitting down. that gives me, give or take, the rest of the morning before anyone in this office wonders what the cursor is doing.

the napkin is still in my drawer. one circle. one tiny crescent that represents the meals i ate that she did not also eat. that crescent, i would argue, is where my entire life happens. decision making biases are why i think the crescent is significant and why she thinks the crescent is a rounding error. we are both right. we are both running the bias version that flatters us. the napkin, the crescent, and the rest of the morning are what this post is for.

decision making biases: the systematic shortcuts the brain takes when choosing between options, where the option that flatters the chooser gets a bonus, the option that contradicts the chooser gets reviewed twice, and the option nobody noticed gets quietly skipped. the bias picks first. the reasoning shows up later, in a clean shirt, to explain it.

THE BRAIN. PICKS. THEN. ARGUES.

i need that one early because the rest of this post is going to read like a man defending forty thousand small decisions. it is. some of those decisions were the kind of thing the pillar on confirmation bias already covers, which is the engine room for most of the types of decision making biases people google at 11pm on a sunday. the pillar is the engine. this post is the parts catalogue.

1. decision making biases, brief

the textbook outline goes roughly like this, and i am paraphrasing from memory because the laptop is open to a spreadsheet that is supposed to be my full attention. decision making biases are systematic patterns where the brain prefers what is familiar (status-quo bias), what came first (anchoring), what it just heard (recency), what it already believed (confirmation), what would feel embarrassing to lose (sunk cost), and what flatters its own competence (overconfidence). that’s six. there are more. there are always more. that’s the problem with this research biases in thinking and decision making field — every researcher who shows up coins another one and goes home.

i looked it up briefly, in a window kept small in case carla returns from upstairs sooner than the q3 review tends to release her. there is, i’m fairly sure, a study somewhere — possibly in a serious magazine, possibly written by a man with a beard — that catalogues over a hundred and eighty distinct cognitive biases. one hundred and eighty. that’s not a list. that’s an inventory of human personality, alphabetized.

the practical version, for someone trying to get through a friday, is shorter. the brain has a few favourite ways to skip the homework. the homework, in this metaphor, is reality. reality is what happens to you when you skip the homework. the chicken in my fridge is, in this metaphor, also reality. the chicken does not care about my book learning.

knowing the names of the types of biases in decision making does not, in any practical sense, protect you from them. i know about anchoring. i still bought a yoga mat because the next yoga mat over cost more, and that made the cheaper one feel reasonable, even though the correct number of yoga mats i needed was zero. that is anchoring, performed by a man who knows the word for it, in real time, with a credit card.

2. the decision i made about the seventh microwave

here is a worked example, and the example is not flattering, which is fine — it is also not the post’s job to flatter me. the seventh microwave was a decision made under conditions i would describe, generously, as “compromised”. the sixth microwave had recently failed. there was a fork involved. there was a flash. there was the smell of a small electrical event the building manager would later describe as “concerning, again”.

i needed a replacement. i had options. the cheap one. the mid-range one. the one with the spinning plate that spins. the one with the spinning plate that doesn’t spin, which is, as the hot take collection holds, the correct kind, because the microwave plate doesn’t need to spin. i chose the seventh microwave the way i had chosen the previous six: by which one the supermarket put at eye level on the day i happened to be there at 8:47pm with a frozen meal that had become, philosophically, a problem.

that’s not a decision. that’s availability heuristic, performed by a hungry man. (it’s also confirmation bias — i was confirming, with the purchase, that i was still the kind of person who solves problems quickly. seven microwaves of evidence to the contrary did not enter the chat.) the seventh microwave, plainly, lasted eight months. the eighth microwave is, as of this paragraph, on order. that’s the species. that’s us. we don’t decide. we shop while flinching.

3. the decision about the third yoga mat

the third yoga mat lives under the couch since 2023, possibly evolving, and the decision to acquire it was a masterclass in three biases at once, performed in a single afternoon, with a debit card.

bias one: anchoring. the studio one cost ninety-eight dollars. the second-tier one cost forty. forty felt cheap because i had been looking at ninety-eight for fourteen seconds. forty was, in any reality not anchored to ninety-eight, still forty actual dollars i did not have to spare. moneyball is, fundamentally, about a man who refuses to do this — refuses to anchor to the price tag the league has set on a player and instead asks what the actual numbers say. i am not that man. i am the league.

bias two: optimism. i predicted, with the confidence of a man who has predicted seven microwaves into early graves, that this yoga mat would be the one. the third one. the lucky one. the one i’d actually unroll on a wednesday morning before work and stretch on, like a person whose hamstrings were a project rather than a rumour. the prediction was, in retrospect, the entire transaction. the mat was just the receipt the prediction left behind.

bias three: sunk cost, applied preemptively. i had bought two prior yoga mats. those two were, technically, evidence i was not the yoga-mat type. but my brain reframed them as tuition. tuition for the third yoga mat, which would now succeed because the previous two had paid for the lessons. that is not how money works. that is not how mats work. that is, however, how brains work, when they are tired and have a debit card in hand.

stretch break. carla is still upstairs. somebody on this floor is microwaving fish. i will not investigate.

4. the decisions i did not make, listed

this is, possibly, the more interesting list. the decisions i did not make are the ones that, by not being made, were also a kind of types of cognitive biases in decision making. specifically: status-quo bias. when in doubt, the brain picks “leave it” because “leave it” feels safer than “change it”, even when “leave it” is actively on fire.

i did not, in 2022, change my phone plan. i was paying, by my own honest accounting, twelve dollars a month for things i did not use. that’s one hundred and forty-four dollars a year, gone, because i could not face a fifteen-minute call with a representative who would, within two minutes, ask if i wanted to “keep the line for an additional fee”. i did not.

i did not, in 2023, open the certified letter that arrived in march. the unopened mail pile took the letter and gave it a place in the system. the system is the floor by the door. the system has been operational for nine years. the letter, last i checked, was still there, slightly pinker than the others. i suspect it knows.

i did not, on twelve separate occasions, return a phone call from a number i did not recognise. status-quo bias says: leaving the call unreturned is the safe option. status-quo bias is wrong about this, demonstrably, since one of those calls was probably the dentist confirming a cleaning i then missed and was charged for anyway. but status-quo bias does not learn. status-quo bias only votes, and it votes the same way every time.

and yes — savings accounts are a hobby for the wealthy is, technically, also a decision-making bias. it’s present bias, the brain’s tendency to value the seventy-eight dollars in a coffee budget today over the seventy-eight dollars compounding into a slightly larger seventy-eight by 2061. i agree with my brain. that’s also a bias. (this might be the post, in fact, where i admit that knowing about a bias and agreeing with it anyway is the most distilled version of the species. you’d think this would be a useful insight. i have done nothing with it.)

5. the office air and decision quality

here is something i have been quietly testing, because we are on the third floor and the air on the third floor has its own opinions about what’s a good idea. when the AC is overcooked — which is most days — i make decisions like a man on a sugar low. when the AC is off — which is rare and always the result of an “incident” — i make decisions like a man on a sauna bench. somewhere between, on the four days a year the AC works correctly, i make the kind of decisions a tom-type would make. tom would not call them brave. tom would call them “reasonable”, which is tom’s highest praise.

this is not research biases in thinking and decision making. this is observation, performed badly, by a man who is also the test subject. but the data, such as it is, is unanimous: i sign off on more emails before lunch when the office is at sixty-eight degrees, i open more tabs on a slow afternoon when it’s at seventy-four, and i delete more drafts on a thursday after carla returns from a meeting, regardless of the temperature. the variables are interacting. the experiment is corrupted. the experiment is also my life.

this is also, for what it’s worth, why every meeting could be a 3-line email. half the decisions made in those meetings are made by people whose air-quality budget the building does not respect. the other half are made by people who have already decided and are gathering an audience. the slide deck is decor. the discussion is a vote that has already been counted, performed slowly, in a room with bad lighting, while someone refills coffee in the back.

6. verdict, the bias makes the decision, i sign it

here is where we land, with the q3 review still going on upstairs and the napkin still in my drawer.

decision making biases are not, in the end, things you “overcome”. you don’t overcome the brain. the brain is the management. you can, on a good morning, with coffee, before any meeting on the third floor, briefly notice the bias running and reduce its priority. that’s the whole win. that is, near as i can tell, the entire skill. anybody who claims to be “unbiased” is, with the calmest possible voice, telling you the bias is so deep they no longer detect it. those are the dangerous ones. the rest of us, with our seventh microwaves and our third yoga mats and our pinker certified letters, are at least operating with the headlights on.

if you came in for a list of types of decision making biases and instead got six examples from one man’s drawer, i sympathise. if you want the cleaner pre-school version of the same machinery, the work on how to tell if you are smart sits adjacent and reads better with a second coffee. it asks the same question from the back door.

i scored, on my own four-question retest from saturday, a 2. then a 3. then a 2 again on tuesday, after the seventh microwave decision recurred to me without warning. the score is not the point. the score is the bias being polite.

napkin venn diagram is now in my pocket. carla just emailed: “do you have the q3 carryover figures?” the carryover figures are in tab 23. tab 23 is titled, currently, “yoga mats? probably not”. i will, in approximately four minutes, decide which version of myself to be in the reply. the bias has already decided. i’m just waiting for the press secretary to put on a clean shirt.