Does speeding save time?

[noelbotevera]

A little thought experiment I've had in the back of my mind is whether speeding will get me places faster. I don't enjoy driving on familiar roads (thus I'm not a big fan of commuting, but that's a thread for another time), so unless I'm on vacation, I generally want to get from point A to point B as quickly as possible. How about I try to math it out?

Part 1: Making Some Assumptions
Honestly, I'm not putting much effort and using the best tools here. I'm using Google Maps and GSV to confirm speed limits, and my methods of calculation aren't very exact. Still, it's good practice to lay out some assumptions and develop a methodology.

I'm going to start by picking a route in Google Maps and try to provide some justification for each route. The base time is the time given on Google Maps, and does not obey the "Ironman" assumption (see below) nor traffic lights. Time saved by speeding is calculated by using the route's distance and dividing by the faster speed (MPH over the limit), which will give me units of time.

Basically: Base time - (Distance / Speed) = Saved time

Finally, here's what I'm assuming:

• The base times given by Google assume that you follow the speed limit.
• No traffic. If I'm able to speed then I generally don't have to deal with traffic.
• No construction, and therefore no detours due to construction.
• No sightseeing; I'm following the gosh darn GPS.
• I will consider traffic lights, which will add a flat 1 minute to travel time. For the sake of simplicity, let's say I stop at half of the traffic lights along the route and round up fractions (so 13 lights = 7 lights stopped at).
• I'm not an insane driver. The "speeding speed" will be something like 5-10 MPH above the posted speed limit. If there are segments that are egregiously low, I'll call them out. Sometimes these are obvious speed traps; unfortunately, you still have to obey these unless you want to get ticketed and slowed down even more.
• I'm not an ironman. I have to stop for food and/or drink every 2.5 hours, and I have to use the bathroom. A stop will add a flat 20 minutes to travel time.

Part 2: Some Example Cases
Case A: Short Distance
As a junior at Penn State, if I want to catch the latest blockbusters with friends, I need to drive out to a theater outside of State College. I might also want to visit this Walmart because the other Walmarts don't have things in stock that I need (say, furniture or kitchen supplies...yes, you can cook on a hot plate).

Here's the route. The road segments have the following speed limits:

• Roads on campus (0.2 miles): 25 MPH
• Atherton Street and Park Avenue (up to Porter Road) (1.2 miles): 35 MPH
• Park Avenue beyond Porter Road (2.2 miles): 45 MPH; can comfortably go 55 MPH here
• I-99/US 220 (1.7 miles): 65 MPH; can comfortably go 75 MPH here
• Shiloh Road (0.8 miles): 45 MPH

I'm fudging things a bit. Obviously I won't go 45 MPH in a parking lot, and there's some ramps involved. There are 12 traffic lights on this route, so following my assumptions, I'll expect to spend 6 minutes waiting at lights.

I get a base time of 11 minutes to travel 6.1 miles. By speeding (adding 5 MPH except otherwise mentioned), this route will take me 7 minutes without considering traffic lights (so 4 minutes are saved). But the amount of time added by traffic lights is 6 (!) minutes, so I'm expecting this trip to take around 13 minutes.

Total time saved: -2 minutes

Case B: A Medium Distance
Yeesh. Clearly I lost a lot of time waiting at traffic lights. Perhaps I could pick a better route that involves less traffic lights, but I'd have to shop (doable) or move (not doable) somewhere else. Maybe I could try a longer route with a little more distance, and higher speed limits so I can speed more.

Maybe I want to meet up with a friend at Hersheypark this weekend, and spend the day having an absolute blast. I want to get to the park as early as possible to ride the most rides, so I don't plan on sightseeing. This trip is longer, but not so long that I'll need to stop for gas or even food.

Here's that route. Here are the speed limits of the following segments:

• Roads on campus (0.2 miles): 25 MPH
• Atherton Street (from Curtin Road to W. Branch Road) (2.3 miles): 35 MPH; can comfortably go 45 MPH
• Atherton Street (from W. Branch Road to US 322 merge) (3.1 miles): 45 MPH
• US 322 (from Atherton Street to Milroy) (16.9 miles): 55 MPH; can comfortably go 70 MPH, except for a 2 lane section (7.7 miles) and some gnarly curves around Laurel Creek Reservoir (3.2 miles)
• US 322 (Milroy to US 15) (48.8 miles): 65 MPH; can comfortably go 75 MPH
• US 322 (US 15 to Clarks Ferry Bridge) (1.7 miles): 45 MPH
• US 322 (Clarks Ferry Bridge to I-81) (11.1 miles): 55 MPH; can comfortably go 65 MPH, it's 55 because it's fairly narrow
• I-81 (US 322/exit 67 to PA 39/exit 77) (11.1 miles): 65 MPH; can comfortably go 75 MPH
• PA 39 (I-81 to Hersheypark Drive) (5.7 miles): 45 MPH

This route has 20 traffic lights, so 10 minutes will be spent waiting.

To travel this route (100 miles), I get a base time of 107 minutes. Without considering traffic lights and also speeding, this will instead take me 90 minutes (time saved = 17 minutes). Accounting for traffic lights, this trip is expected to take 100 minutes.

Total time saved: 7 minutes

Case C: A Long(ish) Distance
Okay, let's get a little out there. Now I've decided to spoil myself and take a trip out to New York. Maybe I'll catch a Knicks game, visit some friends, and eat $16 halal cart food. Or visit the glorious Chambers Street Station. Furthermore, I've heard hotels in Queens are cheaper than staying in Manhattan, and if I'm lucky entering New York won't be so painful.

Here's that route. (I just picked a random hotel in Queens. I have stayed in Astoria in real life; it's one of my favorite neighborhoods mostly due to the Hell Gate Bridge.) The road segments have the following speed limits:

• Roads on campus (0.2 miles): 25 MPH
• Atherton Street and Park Avenue (up to Porter Road) (1.2 miles): 35 MPH
• Park Avenue beyond Porter Road (2.2 miles): 45 MPH; can comfortably go 55 MPH here
• I-99/US 220 (from Park Avenue/exit 73 to I-80) (10.8 miles): 65 MPH; can comfortably go 75 MPH here
• I-80 (from I-99/US 220/exit 161 to I-380/exit 293) (133.6 miles): 70 MPH
• I-80 (from I-380/exit 293 to US 209/exit 304) (12 miles): 65 MPH; 75 MPH is possible, but...
• I-80 (from US 209/exit 304 to US 46/exit 4) (10.6 miles): 55 MPH; I-80 through Stroudsburg and the Delaware Water Gap is old and narrow, so the 65 zone is probably a warning
• I-80 (from US 46/exit 4 to US 46/NJ 23/exit 53) (49.1 miles): 65 MPH
• I-80 (from US 46/NJ 23/exit 53 to I-95/exit 68) (15.6 miles): 55 MPH, but let's face it, most people do 70+
• I-95 (from I-80/exit 68 to I-87/exit 3) (7.1 miles): 55 MPH, could do 65 in no traffic
• I-87 (from I-95/exit 7 to I-278/exit 1) (3.5 miles): 55 MPH
• I-278 (from I-87/exit 47 to 31st Street/exit 45) (2.7 miles): 55 MPH
• NYC city streets (1.9 miles): 25 MPH

This route has 24 traffic signals, and requires me to stop to rest once.

The base time for this route (246 miles) is 232 minutes. By speeding, this trip will take me 213 minutes (so the amount of time saved is 19 minutes). Accounting for traffic lights (12 minutes) and rest (20 minutes), this trip is expected to take 245 minutes.

Total time saved: -13 minutes

Part 3: What Does It All Mean?
Uhh, speeding doesn't save time, but don't take my word for it. I'll start by pointing out several issues with my methods:

• You can spend more than 1 minute and stop at more than half of the traffic lights on a route.
• Not assuming traffic, but I think if you're able to speed, you're probably not in traffic.
• Not dedicating time to sightseeing. But if you're sightseeing, you're probably not in a rush to get anywhere and thus not likely to speed.
• Using only Google Maps and relying on imperfect research. I've done these trips myself, especially the shorter ones, but I can't be bothered to drive out over 200 miles to actually test my theory.
• All of the examples listed are on the East Coast. Obviously this is location dependent and I'm sure travelling 200 miles on an interstate posted at 80 MPH in Utah is different than 200 miles at 80 MPH on an interstate in the East.

Still though, what can be gleaned is that there's a sweet spot where speeding...probably saves time? If you don't have to stop on a trip, and there aren't many traffic lights, then yeah you can shave 5 or 10 minutes off of your trip. Probably useful for a medium range trip, but you could also just leave earlier if you have to travel an hour for an appointment or something.

Maybe speeding just makes us feel better about ourselves.

[Max Rockatansky]

Depends on the context, probably not so much in a dense urban area with traffic lights.  Out in the county and with significant distance speeding can definitely save you time.  The average estimate I get for heading home is 48 minutes (sometimes gets as low as 45) with traffic is normal.  I easily get there between 38-42 minutes (sometimes as low as 36 minutes) on a regular basis going 7 MPH or by following fast cars.

[noelbotevera]

Depends on the context, probably not so much in a dense urban area with traffic lights.  Out in the county and with significant distance speeding can definitely save you time.  The average estimate I get for heading home is 48 minutes (sometimes gets as low as 45) with traffic is normal.  I easily get there between 38-42 minutes (sometimes as low as 36 minutes) on a regular basis going 7 MPH or by following fast cars.

Eh, that was the point I was trying to make. Yeah, by driving faster on a relatively long trip (so that you don't have to stop), you'll save around ten minutes. Does it really matter? I guess being home ten minutes sooner means I can eat dinner a little earlier.

Speeding on a longer trip (over 3 hours in length) will save you more time (maybe 20, 30, 40 minutes), but if I'm driving from State College to Chicago I'm still going to arrive in the evening no matter how you slice it.

And if there's traffic then all bets about saving time are off.

[citrus]

The assumption that the Google Maps times follow speed limits is probably wrong. As I understand it, the times are based on real-world data, so if most people speed, then Google's time will probably involve speeding too.

[Max Rockatansky]

Seemingly it calculates the amount of time anticipated to be stopped at traffic lights.  Some of the mountain road time estimates around me definitely are on the pessimistic side and appear to be calculating way below the speed limit.

[vdeane]

The math seems funny to me.  On the short range trip, you claim speeding shaves four minutes of time in the middle of the calculation, but then you add in the lights and end up claiming that speeding adds two minutes of time in the end.  Then you do the same thing for the other scenarios.  Aren't you comparing applies to oranges when you do that?  Light time is one of the things that varies in a trip, but I can't imagine lights causing a trip to take longer speeding than it would religiously following the speed limit.  Worst case scenario is that you sit at a light that would have turned green by the time you got to it if you didn't speed.  Best case scenario is you make a light that you otherwise wouldn't have.

Also, Google Maps does assume speeding by a small amount, at least in the northeast.  Take the Thruway for instance: at 65 mph, going 65 miles should take 60 minutes, but Google says it takes 56.  I find that when I go my usual 5-7 over, Google's time tends to be right on (and I don't factor additional time for lights, so Google somehow compensates for them), but if I go the speed limit, it takes longer than Google says, especially for multi-hour trips.

In any case, maybe you'll value your time more once you're out of college and time shifts from something you have in abundance to something that's quite scarce (sure, it might not feel abundant now, but once you have a "9-5" office job, it will).  Wanting to be at places by certain times adds another dimension to it too.  For me, I want to get where I'm going in time to watch the evening news (both local and national; if I'm traveling elsewhere, I only care about the 6-7 block and can sometimes wiggle it depending on the media market, but if I'm going to be home, I want to keep my usual routine of starting at 5).  As such, arriving a half hour later would be a big deal, especially since leaving earlier isn't something I'm good at.

Not to mention that driving exactly the speed limit just feels like I'm crawling, especially on interstates.

[noelbotevera]

The math seems funny to me.  On the short range trip, you claim speeding shaves four minutes of time in the middle of the calculation, but then you add in the lights and end up claiming that speeding adds two minutes of time in the end.  Then you do the same thing for the other scenarios.  Aren't you comparing applies to oranges when you do that?  Light time is one of the things that varies in a trip, but I can't imagine lights causing a trip to take longer speeding than it would religiously following the speed limit.  Worst case scenario is that you sit at a light that would have turned green by the time you got to it if you didn't speed.  Best case scenario is you make a light that you otherwise wouldn't have.

Good catch. I'm being handwavey about traffic lights (mostly because I'm not smart enough to figure out how many and how long people will spend at a light (and further complicated by the fact that turning left/right will take longer than going straight)), but my intention is that Google's time assumes a perfect world (speed limit but no traffic lights). Then I calculate time saved by speeding (speeding + no traffic lights) and add in the wrinkle of the real world that you'll probably wait at a traffic light (so total time = base time w/ speeding + other stuff (traffic lights, rest, construction, etc.))

I realize it is a bad faith comparison to add traffic lights, but I thought it'd be disingenuous to outright ignore them. Should I just focus on the time saved and remove the time added by traffic lights?

In any case, maybe you'll value your time more once you're out of college and time shifts from something you have in abundance to something that's quite scarce (sure, it might not feel abundant now, but once you have a "9-5" office job, it will).

Can't wait. I think commuting longer than thirty minutes to work just means less chances to do stuff throughout the day (or that means an hour of unpaid work every day).

Not to mention that driving exactly the speed limit just feels like I'm crawling, especially on interstates.

True, but I feel that's because everyone these days wants to go above the speed limit to the point of tailgating slow drivers. I don't know why everyone wants to go 75 in a 65, but if nobody will enforce it, then who cares?

[Flint1979]

Google Maps follows average speed I do believe.

[1995hoo]

It's funny, back in the 1990s I used to go 90+ mph on I-85, yet I never managed to make the 260-mile drive home to Fairfax County in less than four hours, no matter how fast I went on that road. Invariably I'd hit traffic somewhere and lose whatever time I'd gained.

Now, with that said, I can think of times when speeding did save time (such as one New Year's Eve when I drove 650 miles in under nine hours and a trip to Atlanta once when I went 560 miles in seven hours).

But you know what, nowadays I just can't be bothered to try to go that fast very often anymore, and I find overall it makes for a more relaxing drive, especially because I don't care if I see a cop. Everyone else is always slamming on the brakes in a panic. I'm also pretty strict about lane discipline, but the side effect of that is that if I'm in the right lane going 70 mph in a 70-mph zone, or 65 mph in a 55-mph zone, and you tailgate because you think you're entitled to go faster, I'm not going to give a damn about you and I'm not going to speed up (especially because I probably have my cruise control set).

[Rothman]

Meh.  Just comparing Google Maps directions doesn't paint the whole story.

On long distance trips, what haunts me is where that threshold is where, if you speed too much, you're incurring more stops for gas and thus actually slowing you down.  Haven't figured it out yet, since gas mileage is so nuanced and affected by a host of factors.

That said, I beat Google Maps' times regularly on long trips by quite big margins.

[jeffandnicole]

Where you're going wrong is adding in the 1 minute at every other red light and the 20 minute breaks when you speed, but not adding them in when you're not speeding.  These factors would be the same regardless if you're going the speed limit or speeding at the speeds you're estimating.

To use your first example:  Your basetime is 11 minutes.  By speeding, you'll only need 7 minutes.  Yet, you only added in the traffic light stops when speeding, not when going the speed limit.  Maybe you're driving thru red lights while going the limit?


For me, the only time I give this much thought is when I'm taking a really long trip, such as from NJ to FL.  There, if I drive 74 mph (my general highway speed I tend to go now in 65 & 70 mph areas) vs 78 mph, it's probably costing me about an hour's time (4 miles per hour for 15 hours).  Why am I driving slower?  To save a bit on gas.  I'm not worried about speeding tickets, where my observation consistently shows keeping it below 80 on 65 and 70 mph roads results in no one caring.

[webny99]

Google Maps definitely factors in extra time for traffic lights. I'm not sure exactly how the algorithm works and I don't think it's as much as a minute per light, but you can reliably save a minute or two by making several green lights in a row.

Last time I took NY 104 to Greece, I only stopped for two lights out of 20+ between the Veterans Memorial Bridge and Elmgrove Rd, and that improved my ETA by 4 minutes over a span of 12 minutes, or about 1 minute saved for every 3 traveled. That's with a very high concentration of lights though, speeding as necessary to reduce the risk of the next light changing, and some good luck to top it off.

[Mapmikey]

Speeding cannot be slower in a suburban setting with stoplights, unless the stoplights are triggered to turn red for a vehicle going over a certain speed, as Charlottesville VA used to have on US 29.

But speeding in the same suburban setting with non-triggering lights can only be faster if you can get through some stoplight intersections on green you would not have at the speed limit.  Or if you are going so fast you get to a stoplight an entire cycle sooner (think if you were 'racing' someone going the speed limit - if you get to a light a cycle sooner you both sit through the light but the speeder will get to the destination sooner)

Otherwise, at least in a traffic light dense area, the speeder spends more time at the light than the non-speeder but both leave the light at the same time.  At sporadic traffic light spacing the speeder eventually clears a light that catches the non-speeder.  You have probably observed someone who takes off at every light but you end up right behind them at the next light anyway.

[1995hoo]

Where you're going wrong is adding in the 1 minute at every other red light and the 20 minute breaks when you speed, but not adding them in when you're not speeding.  These factors would be the same regardless if you're going the speed limit or speeding at the speeds you're estimating.

To use your first example:  Your basetime is 11 minutes.  By speeding, you'll only need 7 minutes.  Yet, you only added in the traffic light stops when speeding, not when going the speed limit.  Maybe you're driving thru red lights while going the limit?

...

This is not necessarily the case, but you have to know the road very well to know whether it matters. I can think of a local road where if you go the speed limit, you'll hit a bunch of red lights, but if you go 5 mph over the limit, you stand a good chance of getting through all of them on green. So that's a situation where "speeding" can save time by itself helping you avoid stopping at red lights. But of course even then it's in exact because other things can factor into whether you wind up hitting a red light anyway, such as if someone going slower changes lanes in front of you and causes you to slow momentarily, or your own personal level of comfort with going through on a yellow that's about to go red.

[jeffandnicole]

Google Maps definitely factors in extra time for traffic lights. I'm not sure exactly how the algorithm works and I don't think it's as much as a minute per light, but you can reliably save a minute or two by making several green lights in a row.

Last time I took NY 104 to Greece, I only stopped for two lights out of 20+ between the Veterans Memorial Bridge and Elmgrove Rd, and that improved my ETA by 4 minutes over a span of 12 minutes, or about 1 minute saved for every 3 traveled. That's with a very high concentration of lights though, speeding as necessary to reduce the risk of the next light changing, and some good luck to top it off.

Right, but that's my point: Google Maps is factoring in that time at a light already*. By adding an extra minute per 2 lights manually, that's not overriding GMs estimates; it's in addtion to. As Mapmikey mentioned - speeding may cause someone to spend extra time at a light compared to someone driving slower and maybe timing the lights better, but that would mean the total time for both drives is about the same. The speeding driver isn't going to fall behind the law abiding driver.

* - Not sure what that estimate is either, but I know it's not estimating lights that are known to have long cycle times of red for 3 or 4 minutes.

[kkt]

Since Google Maps is not transparent about how exactly their estimates are arrived at, I would ignore them and  make your own observations.

[Jim]

I would say the big risk for being slower when speeding is the time spent with the friendly officer who pulls you over.  Otherwise the math says the red lights can only slow you down as much as they would if you arrived at them a little later and waited a little less.  Catch a few green or yellow and you're winning.