31:T612,md`The risk of confusing readers about the ultimate outcome of the race can be seen in the New York Times page reporting results from California's 10th Congressional District. On Wednesday, November 7, the day after polls closed, this is how the page appeared, according to a snapshot taken by the [Internet Archive](https://web.archive.org/web/20181107205804/https://www.nytimes.com/elections/results/california-house-district-10). I've added a red box to emphasize how precincts reporting were described. ![Before](https://gist.githubusercontent.com/palewire/d653820292a4825f647f4a32b1fe118c/raw/05ed45fc82ea95f864623af7f1b2f1aad0f035d1/before.png) [A week later](https://web.archive.org/web/20181114171358/https://www.nytimes.com/elections/results/california-house-district-10), after 75,000 more provisional and vote-by-mail ballots had been counted, the result reversed. The precincts reporting figure, which I've emphasized again, is unchanged. Some observers believe this contributes to public confusion, or at least is an imperfection in the report. ![After](https://gist.github.com/palewire/d653820292a4825f647f4a32b1fe118c/raw/05ed45fc82ea95f864623af7f1b2f1aad0f035d1/after.png) This display is commonplace, including on the competing tables I personally produced [at the Los Angeles Times](http://www.latimes.com/projects/la-pol-na-us-general-election-results-2018/). I point it out on the New York Times because its pages are well preserved by the Internet Archive, and I consider the NYT to be the gold standard in information graphics.`32:T927,{ var h = html`

Candidate	Party	Votes	Pct.
Jeff Denham*	Republican	56,701	50.6%
Josh Harder	Democrat	55,414	49.4%

112,115 votes counted, 100% of polling precincts reporting (478 of 478). The vote totals will continue to change as mail-in ballots are counted.

There are 329,483 registered voters in the district and the turnout in the last midterm election in 2014 was 52%, or 164,741 votes. California law requires the election result to be certified by December 1.

If you want, we'll let you know when this race updates.

* Incumbent

` var a = h.querySelector('#turn-on-notifications') a.addEventListener('click', e => { window.Notification.requestPermission() }) return h }33:T7dc,html` `34:[] 2f:{"title":"Election results challenge: Precincts reporting","cells":[{"id":0,"value":"md`# Election results challenge: Precincts reporting\n\nA week after the 2018 midterm elections, U.S. officials are still counting ballots. A handful of close races remain undecided.\n\nIn states like California, the increasing number of provisional and mail-in ballots tallied following Election Day has made the outcome after the initial round of results less certain. News consumers, used to decisive calls after polls close, need to be informed that tight races will now take days or weeks to be finalized.\n\n[Voices](https://twitter.com/B_M_Finnigan/status/1062591693722017794) in the news media are calling for a reform of how television broadcasters and online publishers commonly describe \"precincts reporting.\" Typically this number refers exclusively to votes collected at polling places and counted on Election Day. It entirely ignores the pile of votes, increasing each year, that must be counted in the days that follow.`","pinned":false,"hidden":false,"mode":"ojs"},{"id":48,"value":"$31","pinned":false,"hidden":false,"mode":"ojs"},{"id":50,"value":"md`### Got a better idea? Here's you chance to prove it\n\nRather than complain, let's come up with something better. I've created a simplified version of the New York Times table from the day after the election below using the excellent tools available here at Observable.\n\n**Your challenge:** Fork this page, [use Observable's new merge feature](https://beta.observablehq.com/@observablehq/fork-share-merge), and suggest some changes. The best submission will earn a free lunch with the Los Angeles Times Data Desk at the [upcoming NICAR conference in Newport Beach](https://www.ire.org/events-and-training/event/3433/). \n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":115,"value":"md`### My idea`","pinned":false,"hidden":false,"mode":"ojs"},{"id":60,"value":"$32","pinned":false,"hidden":false,"mode":"ojs"},{"id":182,"value":"md`Basically, there's just _more_ information to try and put the complexity of elections/vote counting in context. I find it confusing that the percent reported number is generally a reference to the in-person polling places, but an election is also mail in ballots. This, of course like all aspects of an election, varies state by state. In addition, I wanted to make sure the reader would have a mental model of the total potential votes and when the election absolutely must end. Historical context is always welcome, so I threw in the last similar (midterm or general) election's turnout race.\n\nBut I also don't like that election results encourage a person to just sit on the page all night (or for a few weeks as we've seen in recent contests), especially when we have new-ish web tools at our disposal such as push notifications. This way a user can register that a race is important to them and we can make sure we alert them when the situation on the ground changes.`","pinned":false,"hidden":false,"mode":"ojs"},{"id":62,"value":"$33","pinned":false,"hidden":false,"mode":"ojs"}],"files":"$W34","stdlib":"1","titling":"auto"}