Hi, Sha from Trulia here. I worked on some of the frontend aspects of this visualization and would be happy to answer any questions. We're pulling in OSM road data and published GTFS files for public transit routing. I just linked to SF but you can also search for other cities:
Is this the absolute worst-case commute times? It never takes me that long to commute into/out of miami, or baltimore, or fort lauderdale. More like half the time.
DC actually seemed like it went the opposite way as it usually takes an hour and a half to get from VA through the beltway into DC while the map showed only an hour for as far out as Dale City. A half hour (or more) from Petworth into Downtown during rush hour is fair, since DC has probably the slowest crappiest drivers in the nation.
Updated theory: these commute times might not be based on actual driver data, it's probably a very rough estimation based on distance, travel time, speed limit, and population. It needs to be weighted based on the curve of traffic based on time and the way locals drive. Boca commuting will probably be slower than Miami commuting on average, unless you're on the highway in which case you're totally screwed in Miami. Not to mention anywhere there's an on-ramp with lots of flow you're going to make a choke point, so times would increase a lot, unless you started after that exit.
Agreed on DC - the places that I plugged in from northern va. (22102, 22213, 22046, etc.) all seemed to be showing the best-case times. Some places that it says are a 5-10 commute are more like 30-45 minutes at rush hour.
Also, not related to the commute times, but the crime rates look strange. At least, it's showing Arlington, VA as almost completely red (very high crime). Maybe the data is right, but that would be surprising for one of the richer, more yuppified areas of the US.
Yeah, looked at Seattle -- what is typically an insane 60m commute (the 520 floating bridge) is a cheerful 20-25m. NYC was also pretty laughable -- traffic really matters outside of sprawl cities with six lane highways masquerading as main streets.
If you are going to measure the distance and use that as a basis -- just report the distance. Otherwise, the estimates based on time will just flip the bozo bit in your customers.
When do you commute? It says Cambridge to Bedford is about an hour, which is true at the start/end of rush hour. It's usually more than that if I get in my car after 3:30pm or before 6pm.
As far as I can tell, this must be the case. I'm just north of Boca Raton, FL and from where I live to work is about 8-10 minutes, not the 25-30 this is saying.
EDIT: The only time it has taken 30 minutes was when I had to wait on two trains.
But go to Miramar and look at how the heat is drawn. Huge swaths of area off the scales, and then north of that, inexplicably, is the heat again, saying that something five miles away is somehow 60 minutes, but nothing in between could be reached in 60 minutes. It just obviously does not represent an accurate commute time for the local area... it's some generic calculation done which they probably assumed was correct when they tested it because the one city they were looking at's traffic pattern correlated with their algorithm.
And i'm not trying to give them a hard time, really. It's a neat visualization. But for me, practically, I couldn't use this to tell actual commute times, and I think that's apparent based on the other comments on here. You want my opinion? They should crowdsource taxi drivers. Who else would have a better idea what the traffic's like at any given time of day?
Times seem inflated in Austin, too. The smaller the time, the greater the error. The ten-minute area should be two or three times larger, but the 45-minute area bears some resemblance to rush-hour reality.
Also, every route seems to be treated the same. In the morning, the road I live on is always empty in one direction and congested in the other. I used to commute against traffic and now go the same way as everybody else, and I could go twice as far in ten minutes as I can now. The numbers will never be accurate until you ground them in real traffic data.
It appears that non-freeway streets seem way off, with freeway travel closer to being accurate.
For example, Beaverton, OR, into Portland. Where I live it's at most, even at rush hour, 15 minutes to one of the two freeways near my house and this is estimating 30-35 minutes. The entire (house -> downtown) calculation is close (about 45 minutes downtown) but only for rush hour. Other than if there was an accident on I-5, there isn't a possibility of taking 50-55 minutes to get from my house to Tigard or Tualatin without spending 30 minutes lounging at Starbucks along the way.
Looking at the data a little more, surface streets seem to simply be an estimate outward radially, with what appears to be very little taken into account otherwise. The freeways seem to be good, educated guess estimates.
With regard to some of the earlier replies saying that the map is inaccurate, it might be worthwhile to use Amazon Turk to run some evaluation on how reliable the spatial data (or your algorithm for calculating travel time, rural vs local, .etc) is.
Sen, that is a great idea. We're definitely trying to get this as accurate as possible (not having a public traffic data source makes this hard), and Turk would go a long way towards answering questions about places we don't have a good sense for (even all these comments are great for us to fine tune our calculations).
Hi, as a person looking for a home in Boston I've used every tool out there. Amongst your competitors I find myself gravitating back to you and Zillow for the superior user interface. I love you for being able to click "search entire map".
However the main problem I have during my real estate search has nothing to do with UI. It seems like for every home I find through the site that I contact a realitor about the home is already "under agreement". If your site could find a way to be more up to date about those things, it would really stand apart.
Thanks - it actually seems close to accurate for me - though I have an approximate 2 hour commute door to door so it's purely guessing on the remainder of the trip. Will the outer boroughs of NYC receive transportation options as well?
Where do you live and work? 4 hours of commute a day seems ... excessive. (I'm guessing, based on your question, that you work in NYC but live in another state, and probably spend most of those two hours sitting on a train, which would make it less, uh, awful-seeming.)
Sadly, I live in Staten Island, still one of the boroughs. I live on the southern end of the island, and work near the Empire State Building. While it is indeed excessive, it's on a nice comfortable tour style bus so I get an extra 2 hour nap time in the morning and can get work done on my way home - and I save more money due to far lower rent.
Great project. Would you please allow the slider to increase to values greater than 60 minutes? Unfortunately some of us poor souls in the Northeast regularly have 2 hour commutes into NYC.
I agree with the other observations here that the timings aren't even in the ballpark. For North Denver Metro area, it is displaying 50-60mins for places that are 25 mins away in rush hour.
Not really applicable to me it seems. I commute across the Twin Cities in Minnesota in about 50 minutes but it wouldn't even calculate my times due to it being "well over 60 minutes" which is the max the filter will apply. How useful can you expect this to be, since commuting times can vary wildly even if you begin your commute 10 minutes earlier?
This isn't a website to tell you your exact commute time. If you want that, try Google Maps.
This is to show you roughly estimated travel times in all directions when you're looking at purchasing real-estate. And I think it works very well for that.
Seems inaccurate for my commuting times to various points in the metro as well. Looking at Broadway running east-west to the north of downtown, its a major thoroughfare that can take you to several major highways but its not being reflected.
Have you considered making a map where the color at each point is the average commuting time from there to all points within a radius? For example: http://www.jefftk.com/news/2010-11-17.html
The maps you've added are awesome when you know where you're looking at, but if you don't yet have a place picked out then knowing where to look is good too.
A naive way to compute the kind of summary map I'm interested in is to just make the color at each point on your summary map be the average color on your map for a specific place. Something smarter could weight points by their desirability (I care more about how fast it is to downtown than to an arbitrary point in a suburb) but the naive version is pretty useful on it's own.
The times are calculated with a simple Djikstra's traversal, implemented in the open-source pgRouting (pgrouting.org)
Our existing local information pages were already using google maps, so we didn't go out of our way to create new base maps just for this visualization.
This is cool, however I notice that your transit commute times are more optimistic that drive times in some places.
For example, from downtown Kirkland, WA. You have the bus being faster than driving, there is no HOV lane that would explain this, my real commute choice is driving (10-15 minutes) or bus (35-40 minutes) but your heat map pretty much inverts those numbers.
For the non-technical, what are you using to create/draw the overlay heat maps? I thought it might be some sort of tile server but it seems to be drawn using JS?
I think it's a very awesome idea, and I'm impressed with the visuals. However it seems that several people on here are claiming wildly inaccurate commute times, and I myself have noticed this as well. I think part of the problem is that it is hard to get a full understanding of what rush hour traffic is like, and that the actual volumes of traffic and slowdown differ depending on locality and road. Sever places, including my home town, have way overblown estimates, such as 40min commute on a road that never takes more than 15. and that's a two way, two lane road. However City driving can be much much different, and even dependent on end of day vs. morning. For example, my DC commute was 40min in the morning (it is reverse rush hour) but my commute home was always a hour and 15, on the same road, reverse commute both times.
edit: and if I could make any other suggestion, I would be to allow me to continue to zoom out. The zoom sticks at a certain point, and it is not zoomed out far enough for me.
As many commentors have pointed out, our time-accuracy is not exactly 100% yet. :)
This is our first iteration on this concept, so we are using some heuristics to try to get times roughly close, and accurate relative to each other. We're also validating the user experience, value as a product, etc.
I'm still investigating a good set for nationwide traffic data, which will help give us a much more accurate picture of real-world commute times. Traffic varies so very much city to city :)
Where I live (Santa Cruz, CA) it doesn't know about a local highway that is virtually stationary every afternoon/evening.
Have you considered partnering with someone like Waze to get anonymised actual trips, which will be far more accurate than pure traffic data (or they could be merged).
Most people are missing some of the most amazing stuff here. First, how the slider works without refreshing the tiles. They have tiled the travel costs http://tiles.trulia.com/commute/time_map/data/driving/37.774... and are drawing these changes without having to refresh the map. That's cool. Secondly, they have used Openstreetmap for calculating travel time. I cannot imagine OSM having a street segment cost that is too accurate, so it was probably derived from pure geometry (think about it... traversing .25miles in downtown SF has a much higher cost than .25 miles on the 5 freeway). If instead, you used a Navteq or Teleatlas dataset for this, I bet the isochrone polygons would come out much more accurate. My hats off to the Trulia team, this is good stuff.
The physicist in me won't think it's cool until there's some monte carlo simulations of commute times with some additional random walk information for alternate routes and stop light entropy, then a log-likelihood fit to find the best fit commute time.
Just kidding, it's alright, it's kind of off for the SF/PA commute though, by nearly a factor of two for many places in SF, unless you're doing absolute worst case scenario every day of the week.
If it was accurate, it would be very helpful if, say, I got a new job, or was looking to relocate areas for a job, I could pick out neighborhoods I'd want to live in based off of commute times to my new workplace. And if I knew where I thought I wanted to live, I could see if it would be too far to work.
or if I was on travel, and had a meeting at a client site at point a. but a hotel near my companies office at point b, I could see the travel times, and maybe even go for a hotel closer to point a (or hotels near point a are too expensive, how far would cheaper hotels be for my commute) .
Perhaps a bit harsh, but I have to agree on the usefulness factor. Obviously the UI looks excellent, but it doesn't matter because the data is wildly off for both areas that I've commuted and seems to be a widespread issue based on other comments.
They probably should have focused on getting better data before launching. Garbage in, garbage out I guess. I hope they can improve the integrity of the data because it has a lot of potential.
If someone is looking to relocate to a new (and thus unfamiliar) location and they need to figure out how far they would comfortably be willing to live based on commute times vs. cost of living.
Hey Sha -- I can't speak for the rest of your cities, but Tampa, FL is VERY wrong. I drug the end point to downtown St. Petersburg and the times given appear to be almost double what it truly takes to commute, over a certain threshold. Under 20 minutes seems to be accurate, but it starts trending towards 30, 40, 60 minutes too quickly.
I'll tell you what, though -- great UX and visualization. Very plain to read and utilize the slider functionality. I suggest modifying the legend to be more explicit about what the color bands mean -- perhaps labeling each color, for instance.
Looks nice, great UI, but I think the Minneapolis / St. Paul numbers are a tad bit off. It doesn't seem to show the big changes in time as you cross choke points (394 -> 94 for example).
I started writing a Chrome extension using the API that you could apply to any existing Google Maps v3 map (such as AirBnB), but I've abandoned it since 1) I was disappointed that quite a few sites with Google maps are still using older versions (such as Padmapper) and 2) I lack time. I believe the Chrome extension portion is working, but it needs a UI overlayed on the map for specifying locations of interest.
If anyone is interested in working on this, send me a message and I'll post what I have to github.
You can also enter more than one commute location. So you can find all places that are within 20 minutes of your office AND 30 minutes of your girlfriend's house.
The other tabs have some pretty cool filters, such as proximity to public transit, grocery stores, etc.
They also have heatmaps showing how "walkable" a particular neighborhood is, i.e. how much you can get your daily errands done without needing a car:
Question: Is the pointer the start location or the end location of the commute? It wasn't immediately obvious to me and in many areas it makes a huge difference. It seems like it's just pulling worst case "in traffic" drive times right now.
Also is it possible to adjust the time of day that one commutes? (since many may work shifts off rush hour and encounter little/no traffic or strange/intermittent transit schedules)
On the large scale, you get pretty much the same map whether you're calculating 'to' the marker or 'from' the marker. The only exception is if you plan on commuting in the opposite direction of rush hour traffic, both ways.
i.e., if you think about commuting to work in the morning, the heatmap will show you the times it takes to drive from any given location to the marker. If you're thinking about driving home in the evening, you can think of the time from the marker to any given location.
on the macro scale, most streets aren't one-way, and commute times tend to be symmetric.
The pointer should be the start location - so drag that to where you live - then from there you can see the estimated commute time based on the heat map.
Some of the data is wrong, e.g. it takes me 30 mins to go to my office and Trulia is saying it's 60 minutes (miami).
But, the UI is great, I just wish they had better data, seems like it's just calculated with distance / avg mph, traffic in this problem has a lot to do to get an accurate calculation.
It might be worth doing "rush hour" and "non-rush hour". It takes me 48 minutes door to door from downtown Oakland to Mountain View, going 80 on I-880 and 70-75 on 84 and 101 (i.e. flow of traffic, and at the speed where you won't get a speeding ticket even for going past a cop, especially if you slow down slightly on seeing one..).
It takes 1.5-2h during rush hour on the same route if anything goes wrong (50-75% of the time there is at least one accident or construction).
Both numbers are meaningful, but distinct.
Transit is the same way, in the other direction -- during peak commute hours, trains are frequent and/or synced. During off-hours, you can be waiting 30-59 minutes for a connection, 1-2 times.
I work in Minneapolis and my commute is about 30-40% (15-20 min) less than what this site says. Neat idea, but certainly needs improvement for broad real-world use.
Imagine stating the location of your work and the cost of your time in $/hour (whatever you assign to it), and the size of housing unit you want. The, you would see a map colored based on the delta between the cost of your commute (via either personal vehicle or public transit) and the cost of housing in various areas. I bet that, for most urban dwellers, there would be a few surprising points on the map.
Yeah, I'm seeing consistent ~3x times in my location outside Seattle (a rural, no traffic sort of place). Within Seattle, well, Downtown to Magnolia doesn't take an hour.
It really looks algorithmic based on street size, assuming some specific congestion level. I'm not seeing any indication of some of the known congestion points that I used to have to deal with.
Thinking out loud: In an attempt to vastly improve the data, could a mobile app be written that can "detect" if you are driving (exact location, speed, alignment/frequency to your past paths, etc), in such a way that the data could be automatically collected and crowdsourced?
Fire up the app, let it run in background, and it logs everything passively and adds it to global data set. Feasible?
Forgive me if I've heard wrong or if this is just wishful thinking, but doesn't Google do this with Navigation?
I believe there's an option to view a traffic layer and from what I've seen in my commutes, it's fairly accurate for gauging traffic density. It should also be capable of calculate travel time based upon crowdsourced traffic information from Android users.
They do yeah. On Android there's even a widget that you set with your destination and it'll tell you the journey time. Would be great to see their live traffic data used for this journey time metric across the world. I'm moving interstate here in Australia, would certainly help narrow down choosing suburbs.
I saw something very similar last year. But I don't a clue neither on the website nor the product name. The only thing I remember is that it was a German guy (I guess!). Could you (Trulia) please let us know what/who motivated this work?
I've read it as "Interactive visualization of commute times for all US citizens." the scary thing is that in another ten year this may well be reality, given the path we're on.
You have support for driving and public transit but not for bikes? Seems like you could get some rough estimates by using elevation variance as a multiplier for driving time.
Looking at this reminds me how glad I am of the fact I get to telecommute... and the fact I only put 2900 miles on my truck last year. Very nice visuals though!
This is very optimistic to the actual commute times in Atlanta. They are off by a factor of 2 or 3. Not sure as to the cause but the vector mapping in Atlanta is terrible, so is most likely routing problems or the like.
http://www.trulia.com/local#commute/new-york-ny
http://www.trulia.com/local#commute/miami-fl
http://www.trulia.com/local#commute/austin-tx