Althouse: "The slower is faster (SIF) effect occurs when a system performs worse as its components try to do better."

June 12, 2019

"The slower is faster (SIF) effect occurs when a system performs worse as its components try to do better."

"Thus, a moderate individual efficiency actually leads to a better systemic performance. The SIF effect takes place in a variety of phenomena. We review studies and examples of the SIF effect in pedestrian dynamics, vehicle traffic, traffic light control, logistics, public transport, social dynamics, ecological systems, and adaptation. Drawing on these examples, we generalize common features of the SIF effect and suggest possible future lines of research."

The abstract for "When slower is faster" by Carlos Gershenson and Dirk Helbing, which I found after saying and looking up the phrase "slower is faster." I didn't think I could be the first person to use that phrase, but I did arrive at it on my own, after having a great experience getting something done incredibly quickly by setting out to do it quite slowly.

It's like "less is more." There must be a hundred phrases in that pattern, and I don't mean phrases that are supposed to be understood as a bad — like Orwell's “War is peace" and "Freedom is slavery." I mean phrases that express a good and insightful concept.

56 comments:

Achilles said...: Don't need a study for this.

"Slow is smooth smooth is fast."

Every grunt knows this.; 6/12/19, 12:08 PM
Fen said...: It's a common archery training tool for our speed rounds. People who tried to rush it end up making their shot pattern worse and actually get off fewer shots.

"Slow is Smooth. Smooth is Fast"; 6/12/19, 12:08 PM
Fen said...: Damnit Achilles ninja'd me! But yes, my Primary Marksmanship Instructor (PMI) in boot camp taught me this. I suspect Achilles as well.; 6/12/19, 12:09 PM
Fen said...: This comment has been removed by the author.; 6/12/19, 12:10 PM
JRoberts said...: I always tell my team to focus on accuracy. Speed comes with familiarity.; 6/12/19, 12:11 PM
gbarto said...: A stitch in time saves nine.

I believe it is the distinction between efficiency and effectiveness.; 6/12/19, 12:13 PM
Anonymous said...: Not quite the same pattern, but festina lente, "make haste slowly".; 6/12/19, 12:15 PM
Richard said...: Aesop had a fable about this: "The Tortoise and the Hare."; 6/12/19, 12:18 PM
Achilles said...: Fen said...
Damnit Achilles ninja'd me! But yes, my Primary Marksmanship Instructor (PMI) in boot camp taught me this. I suspect Achilles as well.

For Pretty much everything having to do with the POS M16 they gave us in Basic. Shooting it. Stripping it. Cleaning it.

Came in particularly handy when they started "introducing stress."

It makes total sense in actual combat situations where you have to get control of yourself during an adrenaline release.

It is so obvious to grunts. Once made clear it just makes sense.

My guess is this is taught in every trade in some way.; 6/12/19, 12:18 PM
Fernandinande said...: How fast should an athlete run a race? If she goes too fast, she will burn out and become tired before finishing. If she runs conservatively, she will not get tired, but will not make her best time. To minimize her race time, she has to go as fast as possible but without burning out. If she goes faster, she will still not be as fast as a man.; 6/12/19, 12:20 PM
Fen said...: I was a line marshal for an interesting archery station that demonstrated this.

It was 30 a yard long line with ten targets, each set up at a 20 yard range, about every 3 yards along the line. Competitors had to move along the 30 yard line, set up at each of the 10 positions, take one shot, then move down the line to the next target and repeat.

But only had 60 seconds to complete the course. Unshot targets would count for zero points.

I say "only", but that was the trap. 60 seconds was more than enough time to walk through the line and carefully take each shot.

But every contestant rushed through the line, rushed their shot, hit an average of 4 of the 10 targets. The one "slow is smooth" guy didn't make it to the last 2 targets, but scored 8 of 8 on the others. He won, by almost double the average score.

It was interesting to see how the perception of not having enough time rattled veteran archers who should have known better.; 6/12/19, 12:24 PM
Ann Althouse said...: I find if I rush, I get clumsy and start dropping things, which makes me even less effective. I lose more time that way than I could have saved by speeding up.

It seems related to the fact (I think it's a fact) that you can't read faster by pushing yourself to read faster, you'll just screw up your reading and have to go back (or simply fail to really read it). Slower reading is reading keyed to your level of comprehension, and you build your power of comprehension, which is the only way to read fast.; 6/12/19, 12:28 PM
Anonymous said...: All of our examples so far express the same concept. Anybody think of any that have to do with something beside speed v. efficiency?

I can think of some phenomena that seem like they could be the basis for sayings/insights in this "antonym" pattern, but I can't think of any sayings offhand.; 6/12/19, 12:32 PM
Hagar said...: Grab the ball and run with it!
It is the speed that counts; not the direction!; 6/12/19, 12:33 PM
Fen said...: I find if I rush, I get clumsy and start dropping things, which makes me even less effective. I lose more time that way than I could have saved by speeding up.

Exactly. Worse, most people actually panic at the recognition their speed had made them less effective, and they start to bumblefuck their way though it even less effectively. It's often a double whammy.

I've left archery lines with a goose egg (0 points) after spazzing and firing off double the shots of my opponents that were so wild they didn't even hit paper. It's humiliating. Especially when you know better.; 6/12/19, 12:33 PM
bagoh20 said...: Peak efficiency is what you get when you are in the zone. It's when you are enjoying the task that you do it best and fastest. Which is why I always try to encourage people at work to: "Have fun with it. It's a game. Just try to win."; 6/12/19, 12:36 PM
Fen said...: everything having to do with the POS M16 they gave us in Basic.

Do you remember how awkward and heavy it felt first day they assigned one to you?

After the first year it felt like an extension of your right arm? LOL; 6/12/19, 12:40 PM
Big Mike said...: Basic training is one place where the mantra works, but it applies to any form of competitive shooting that has a time-to-complete component to the scoring. Slow is smooth and smooth is fast. Control your breathing, get your sight picture, squeeze the trigger, be surprised when you hear the gunshot. Next target.; 6/12/19, 12:41 PM
Caligula said...: Local optimization degrades global optimization? If the subject were vehicular traffic, the explanations for why would seem obvious.; 6/12/19, 12:41 PM
Tommy Duncan said...: My father never appeared to be in a hurry when doing projects around the house, but he got a lot done. He was very organized, very focused and could see many steps ahead. He hated wasted effort and regarded his work product as a reflection of his character.; 6/12/19, 12:47 PM
Ignorance is Bliss said...: Back when I was single, I tried to establish the proposition that smaller is bigger, with limited success...; 6/12/19, 12:49 PM
Fernandinande said...: It's like how the coyote in those cartoons caught the road-runner by strapping himself onto an ACME rocket and then blasting himself into a fake tunnel entrance painted on a cliff.; 6/12/19, 12:53 PM
gilbar said...: as Gollem said... "More Haste, Less Speed"; 6/12/19, 12:55 PM
Brian said...: See the Wikipedia entry for "Flow"

I highly recommend "Flow: The Psychology of Optimal Experience" by Mihály Csíkszentmihályi.; 6/12/19, 1:01 PM
gilbar said...: if you're drag racing; a great way to lose, is to go FULL POWER right off the line. The crowds will appreciate the clouds of smoke as you slowly move up the track. (wheel slip is Bad (well, Excessive wheel slip is BAD!)); 6/12/19, 1:10 PM
tcrosse said...: Slower! Slower! She cried....; 6/12/19, 1:12 PM
PM said...: Bad is good.; 6/12/19, 1:18 PM
Leland said...: In systems engineering, what Achilles and Fen noted is closer to reality. Instead of smooth, it is the elimination of variation. Variation is what truly slows the system, even if the variant is otherwise faster. The variant is disruptive of the rest of the system, and that causes a general slowdown.; 6/12/19, 1:34 PM
Ken B said...: I have several aphorisms that I apply to software development (which was my job). One is: Theft is good.
The best way to do something is to not do it.

But the wisest aphorism is from Donald Knuth, and everyone in software should repeat it daily: Premature optimization is the root of all evil.; 6/12/19, 1:37 PM
Stephen said...: Related: measure twice, cut once.; 6/12/19, 1:47 PM
n.n said...: Equal, complementary, and synchronized is a well established imperative for human, computer, and project fitness.; 6/12/19, 2:33 PM
Dave said...: I think of Deming and TQM in that regard. Mistakes are very costly to large operations. For example, I had a Nissan Sentra that had a systemic problem within the engine: piston slap in the number three cylinder. When I started doing research on it, I found out that Nissan was making the engine in a new plant and all the engines coming from there had that problem. My car was under warranty so Nissan replaced the engine. As you can imagine, they did not go back with the same engine but changed it to a new, but older model.

Deming asserted that quality was not the job of inspectors at the end of the process but with the worker performing the assembly. That worker in Toyota, the lowest person in the plant, has the power to shut down the assembly line and summon to his workspace the highest ranking person with executive authority in the plant to examine the problem directly on the line.

What is the highest quality automobile you can buy? In my experience that is a Toyota. Quality slows you down up front, but in the long haul it speeds you past competitors who have more recalls.; 6/12/19, 3:04 PM
reader said...: Signs posted all over the machine shop floor, “Right the first time, every time”.

Back in the days of typing tests you just had to relax into it and let your fingers fly.; 6/12/19, 3:06 PM
wildswan said...: Here's two;.

Less is more.
This had to do with writing while working in the real world.

Don't get it right, get it written.
This was college term paper advice; 6/12/19, 3:16 PM
John Clifford said...: "Slow is smooth, smooth is fast" is a quote that Jeff Cooper attributed to Wyatt Earp. The old gunfighters had another saying: take your time quickly.

I read the referenced SIF article, and wondered why they were asking a question (why does going slower lead to faster?) to which we already know the answer. It's all due to system dynamics. A system is comprised of a multitude of components that contribute to the accomplishing of a larger goal (Deming). It's well-known that local optimization (optimization of a subsystem) leads to global suboptimization, and that suboptimization is due to the creation of excessive WIP that bogs down the system. Take one example: the airline passengers trying to escape by an emergency door. If too many passengers are 'pushy' (aggressively trying to accomplish the goal individually, as per the article), then there is a traffic jam at the exit (too much WIP at a point where it cannot be handled) and no one can escape. If passengers slow down enough to let people through the door, then the constraint to flow is removed everyone can escape.

Really, it comes down to the difference between being busy (utilization) and being effective (productivity). This is the key issue that most corporations struggle with today. Measuring utilization (how many hours X how many people) is well-understood as a first-order approximation of effort, while measuring productivity/throughput is not generally well-understood... yet it is the fundamental measure of a successful business. Of course, the things that matter are generally counterintuitive to 'common sense.' And that is why the people who wonder about SIF and why it works when it shouldn't are puzzled... they are not looking past the first-order effect. That is, in a nutshell, what is wrong with things today. No one looks past first-order effects.; 6/12/19, 5:10 PM
Mike (MJB Wolf) said...: Anyone upthread add this close cousin yet?

“Slow is smooth. Smooth is fast.”; 6/12/19, 5:23 PM
Mike (MJB Wolf) said...: Okay first one did. Ha.; 6/12/19, 5:23 PM
Achilles said...: Fen said...
everything having to do with the POS M16 they gave us in Basic.

Do you remember how awkward and heavy it felt first day they assigned one to you?

After the first year it felt like an extension of your right arm? LOL

It sure made the modified pistol grip m4 with a shorter barrel and forgrip mounted on the lower rail with a modified single point gun sling I deployed with feel manageable if not downright comfortable.

In those shit holes you feel naked without it.; 6/12/19, 5:37 PM
Owen said...: Great comments. To amplify the point about local optimization versus system optimization, think of pacing yourself on a highway, especially one with traffic lights (Delaware Route 13 comes to mind). If you rush to the next light, you may have stop: wastes gas and brakes. If you look ahead and slow down earlier, you can often time the green and maintain an effective higher speed with less energy consumed.

That idea of anticipation is maybe like avoiding premature optimization, which locks you in to a design/build investment and denies you information from the future “play” of the system components.; 6/12/19, 5:59 PM
tcrosse said...: There's never time to do it right, but there's always time to do it over.; 6/12/19, 6:08 PM
Kirk Parker said...: Owen,

There are indeed fancy interactions that are sometimes surprising, but the main point against premature optimization is much humbler: you can't really tell, in advance, where the system will be spending most of its time--and so you don't want to waste any human programmers' time optimizing something that doesn't actually have a major contribution to the speed of the system, and possibly introduce a bug from your fancy or complicated optimization that wasn't there in the more straightforward, supposedly slow algorithm.; 6/12/19, 6:51 PM
Kirk Parker said...: tcrosse @ 6:08pm,

Ah, I see you worked there too!; 6/12/19, 6:53 PM
Sheridan said...: My 82nd Airborne son taught me the "slow smooth - smooth fast" aphorism. A variant I've seen is "slow is steady - steady is smooth - smooth is fast". The mantra is applicable in all endeavors and is particularly helpful when one is getting physically/mentally clumsy due to fatigue or environmental effects. I bet loggers are chanting the mantra continually.; 6/12/19, 6:58 PM
LKReinitz said...: Nobody goes there anymore it's too crowded ~~ Y. Berra
What does not kill me makes me stronger ~~ F. Nietzsche
Appear weak when you are strong, and strong when you are weak. ~~ Sun T.
I know that I know nothing ~~ Socrates (But really Plato)
Known Unknowns ~~ D. Rumsfeld

Off the top of my head.; 6/12/19, 7:12 PM
Howard said...: Tcrosse: that's the Consultant mantra; 6/12/19, 7:57 PM
Howard said...: Punchline: no, son, let's mosey on down there and fuck them all; 6/12/19, 7:59 PM
tcrosse said...: Good
Fast
Cheap
Pick any two.; 6/12/19, 8:31 PM
Assistant Village Idiot said...: GK Chesterton made his writing career around such supposed contradictions or the unexpected, such as "anything that is worth doing is worth doing badly," or "The simplification of anything is always sensational.” He believed that paradox was at the center of deep truths, and repeatedly tied paradox to Christianity as a glory, not a mark against it.; 6/12/19, 8:56 PM
Owen said...: The Arxiv paper is useful. It talks about traffic flow and how it can jam as density, speed, safety spacing, etc interact. I think we all know how one too-fast or too-slow driver can cause a “coagulation.” My own strategy is to create a bubble in all directions and adjust speed to stay in the middle. If I see congestion forming in my neighborhood, I may speed up or drop back; but I gave up fighting it a long time ago. Makes the trip less stressful, maybe faster, rarely slower.

The need to increase spacing with higher speeds (the paper reports this also with crane travel to load ships, and I guess with any system where buffers are used to reduce the “coupling” of system elements) is pretty basic but pretty profound. It reminds me of John Boyd’s OODA loop, where you win the fight by forcing the other guy to overload his cognitive capacity and start reacting to a move you’ve already finished.; 6/12/19, 8:57 PM
Sheridan said...: tcrosse - I worked as project controls manager at a Superfund site in SoCal in early 1990's. ARCO sent an executive to the site to oversee the remediation effort on behalf of the PRP's and his slogan was "cheaper, better, faster". What a joke! It was a Superfund project, not a refinery build. They're still performing remediation nearly 26 years later. Good old EPA!; 6/12/19, 9:02 PM
Now is the winter of our discontent made glorious summer by this son of New York said...: There’s a saying in golf that if you want to hit the ball shorter, swing harder.; 6/13/19, 7:36 AM
Now is the winter of our discontent made glorious summer by this son of New York said...: This comment has been removed by the author.; 6/13/19, 7:37 AM
stlcdr said...: Less haste, more speed.

Seems like this is a 'study' to justify the slow pokes continuing their slow poke behavior, and belittle those who are capable of doing both simple and complex tasks quickly.; 6/13/19, 8:01 AM
stlcdr said...: Oh, and to state the obvious: Slower isn't fast; Fast is fast.; 6/13/19, 8:02 AM
Unknown said...: If everything is a rush job, there are no rush jobs.; 6/13/19, 1:02 PM
Lost My Cookies said...: Top swimmers typically look like they are swimming slower tha they are, mainly because good technique can cancel the outward signs of high cadence. Especially in longer events. Looks slower, is faster.; 6/14/19, 7:10 AM