Two to hospital after Harvard Ave I-5 onramp head-on

Two people were taken to Harborview with serious injuries Tuesday morning after a vehicle collided with another head-on near the northbound I-5 onramp from Harvard Ave E.

Seattle Fire and police responded to the crash just before 10:30 AM in the 2700 block of Harvard Ave E just below Roanoke where northbound vehicular traffic must cross two lanes of southbound traffic to enter the freeway.

Seattle Fire reports that both drivers were in stable condition after being rescued from their vehicles and rushed to the hospital with serious injuries. Animal control was also called to the scene to assist with a dog in one of the crashed vehicles.

The area was closed to traffic during the response. Metro 49 was rerouted during the crash but back on regular service as of noon.

Police are investigating the crash that is under SPD jurisdiction as it happened on a city street, not the onramp.

The Harvard Ave onramp is one of only a few Hill-area feeders for I-5. Like the Harvard onramp, ramps like the northbound entrance at Olive Way include some of the sketchier instances of traffic engineering in the city as drivers must navigate oncoming lanes and, sometimes, crosswalks, to enter the freeway.


SUBSCRIBE TO CHS:  APPRECIATE OUR BREAKING NEWS? SUBSCRIBE HERE TODAY. Subscribers like you help pay for the writers and photographers who provide CHS's daily coverage and help us to swing into action on BREAKING NEWS. Join TODAY to become a subscriber at $1/$5/$10 a month to help CHS provide community news with NO PAYWALL. You can also sign up for a one-time annual payment. Why support CHS? More here.

Subscribe and support CHS Contributors -- $1/$5/$10 per month

31 thoughts on “Two to hospital after Harvard Ave I-5 onramp head-on

    • Timmy73: You lost me at “It is only dangerous if” because obviously there is more than one condition that may add to the inherent danger of such a traffic routing setup. You have no idea what the cause of this instance was. Am I pro tailgating? Obviously not. But it’s not the only potential factor, so save your armchair traffic accident investigation.

    • I’ve driven this onramp many times.

      If you tailgate the vehicle in front of you and think you can enter the ramp right behind them with limited visibility – due to tailgating – and a vehicle is driving Southbound and no time to stop, you will have a head on collision. This isn’t difficult to comprehend, is it?

      The on ramp isn’t dangerous. Careless drivers are dangerous.

    • Max, apparently you’re not familiar with the area. You and Jim should check it out some time. I don’t usually jump to conclusions and we do not know if tailgating was the cause, but this isn’t rocket science.

    • I’m familiar with it. I’ve driven it many times too. It’s a horrendous on-ramp design, regardless of why it happened. It needs a pair of stop signs and/or a flow-control signal. It’s a terrible on-ramp.

  1. Gotta agree with both Justin and JayH here. Entering an on-ramp having to cross oncoming traffic is definitely “one of the sketchier instances of traffic engineering”, and it *is* actually surprising this doesn’t happen more often. For someone to be hit head-on because they were tailgating would mean the car who hit them would likely have just missed hitting the car in front of them too. Either way seems likely the oncoming car would’ve been going pretty fast or they would’ve been able to stop. But since none of us was there, really we’re just speculating. Still pretty shitty traffic engineering. A stop sign or 2 wouldn’t be so tough, would it?

    • A stop sign would slow traffic and so highway engineers wont allow it. Motorist speed and convenience always trumps safety in their dangerous junk science world.

  2. In my experience, to enter the freeway on that on-ramp, you must take a left across Harvard at such an angle that you’re effectively head-on with oncoming traffic for a moment. It’s hard to gauge the time it takes to turn across traffic when you’re at an angle other than 90 degrees.

    Taking a left from eastbound Madison to northbound 19th Ave is another example- I’m surprised more accidents haven’t occurred there. As a pedestrian I avoid crossing the long sides of that intersection. Just like at the I-5 entrance, the geometry can be deceptive.

  3. That IS a very dangerous stretch of roadway. Why? Because vehicles routinely speed there…..especially northbound, prior to entering the onramp….it’s as if they think they are already on the freeway. I cross the street there (at E Edgar St, one block north of Roanoke) on a regular basis, and always am extremely careful because of the speeding vehicles. No one EVER stops to allow me to cross.

  4. I use this ramp 2-3 times per day. The problem really starts at the 3 Roanoke intersections (10th, Harvard, Boylston) which are very inefficient and put people in the mood to take chances when they are then waiting in line to use this ramp. Putting a light at this on-ramp would only put further pressure on those intersections. I’d only support a left turn green arrow (and corresponding red light for southbound traffic) during rush hour (4-7pm). Any red light for northbound traffic is going to further mess up this area.

    What there really needs to be is an additional I-5 northbound ramp. Lakeside is one of the most relaxing, under-utilized arterials in the area. It’s part of my running route. But there really should be an on-ramp to I-5 at Belmont.. The I-5 planners probably thought the Mercer on-ramp could help, but clearly it can’t even handle all the traffic coming from SLU/QA

  5. Live across from here – for all the comments I don’t remember another head on crash. Seems like both cars engaged in a game of chicken and lost. Is it ever worth it ?

  6. The southbound car was traveling way over the speed limit. I entered the I-5 on ramp just in front of the car that got hit head on. Before doing so I watched the southbound vehicle and didnt like what i saw. Given its speed and the icy driving conditions at the time, it seemed more prudent to enter the on ramp than risk getting hit while waiting to turn left. I had sufficient time and turned safely but the vehicle behind me apparently tried to do the same and didnt make it.

    • That is nuts, I’m happy you’re safe. You probably already told the cops you were a witness, but I would do so if you haven’t.

      I have taken this on-ramp many, many times and I am always very careful to have a clear view of both oncoming lanes because I figure my death awaits if I get it wrong. I’m glad the people survived and I hope they are healing and will be OK.

      As a cyclist, fortunately this area can be avoided entirely by taking Broadway Ave. E between Shelby and Roanoke (an option cars don’t have because of the traffic diverter). Or if connecting to Boylston Ave., one can take the trail that cuts under I-5.

    • Your comment is an example of why it is so important for a driver, when turning left, to gauge the speed of any oncoming vehicles. If they’re speeding, stay put until they pass. If they’re driving slowly, go ahead and make the turn.

    • Which is why drivers need to allow ample room between them and the car ahead so they can assess oncoming traffic and decide if it is safe or not to access the ramp. Based on Glenn’s report, it sounds like the driver failed to do this.

  7. To call this ‘sketchy traffic engineering’ is ridiculous, and speaks more to the utter incompetence of Washington drivers, and the failure of the state to properly examine and license those drivers.

    It’s an intersection. Like any intersection, you wait to proceed until it is clear. But god forbid Washington drivers know the basic right of way rules.

    Nope, instead, let’s turn a rote social interaction that should take a tenth of a second to make decisions at intersections, into an opportunity to create an ad hoc sign language at every intersection, to then try to waive each other through in whatever manner seems nifty at the moment, taking 30-50 seconds to sort out.

    It’s not like Seattle drivers who will then putter along merging with I-5 at 35-40 MPH, creating unnecessary congestion and unsafe conditions on the bridge, are in a hurry. So what’s the problem?

    The problem is we live in a formerly bumptious state, that hasn’t grown up. That’s happy to license people in a baby boomer laissez faire fantasy world, where having the same for profit businesses do the training and testing, is magically going to lead to desired outcomes, rather than create perverse incentives to pass everyone and make a state full of shitty drivers.

    Driver competence declined once boomers decided paying for insurance to train drivers in high school was an unbearable drain on their precious property taxes. Never mind, that it wasn’t, and that there’s no reason except Reagan Stockholm Syndrome, that cities or the state couldn’t have self insured. It’s even worse in states like Washington, which have abrogated their responsibility to actually test drivers to standards.

  8. People fly south on harvard, up that hill, all the time (i commute n in the am, s in the pm, usually on the bus). Our entire city needs far more enforcement of the 25MPH speed limit (along with enforcement of people blowing red lights on right turns). You could probably properly time a left green arrow for the ramp with the allison street light, it wouldn’t work perfectly, because of allison street and furman street traffic turning s on harvard, but it could help.

  9. When I moved to Seattle a quarter-century ago, I was shocked that a highway on-ramp had been built like this. It’s a ridiculous and dangerous arrangement. You should never have to move from a standstill to freeway speed in such a short space while darting across oncoming traffic.

    Ironically, the absurdly risky design has likely prevented more accidents. In confusing driving environments lacking clear regulation via signage, drivers often feel uncomfortable and take fewer chances than they otherwise do. But not in this case, apparently.

  10. A genuinely dangerous onramp/intersection requiring far too much correct judgement on the part of all drivers involved. Should be shut down or reconfigured.

    • This was indeed a horrible accident, but is there a history of other accidents at this location? If not, I would hardly call it “genuinely dangerous.” Accidents occur occasionally even at safe intersections.