Is it possible to know how your state representative will vote on a bill before it gets debated? What started off as a casual bet between two reporters has shed light on a little-known tool that can pretty accurately predict such outcomes.
Lawmakers recently debated a measure that would have required Colorado’s ski resorts to make their injury and fatality data public.
The measure pitted one of the state’s most iconic and lucrative industries against safety advocates who say the information would force resorts to become more transparent and accountable to consumers looking to reduce their risks on the slopes.
Before Senate Bill 184 was voted on by a committee in April, Capitol Coverage reporter Scott Franz and KUNC investigative reporter Michael de Yoanna made a bet on the outcome. Scott, moreover, said he could predict how each of the committee’s members would vote.
It’s common for journalists to try to predict the outcome of legislation. Sometimes they share their analysis in their stories, telling their audience whether a bill is expected to pass or fail. One reason for doing this is that reporters and editors must make tough decisions on a daily basis about what to cover — sometimes, it’s important to tell the story of a bill that is expected to fail, but more often, reporters are chasing the story of what’s most likely to pass and become law.
Scott and Michael employed very different methods in their predictions. Here’s what went down.
The players
Scott Franz, Capitol Coverage reporter for Rocky Mountain Community Radio
Before his days roaming the halls of the Capitol building for KUNC and Rocky Mountain Community Radio, Scott covered the ski industry for several years as a print reporter in Steamboat Springs. He sat next to the police scanner and heard paramedics treat injured skiers almost daily. He has covered state politics for public radio since 2018.
Michael de Yoanna, investigative reporter for KUNC
Michael led a two-month investigation into ski area safety in 2019 that found more than 5,600 emergency room visits in a single year for Colorado’s skiers and snowboarders. His reporting uncovered little-known data revealing the extent of ER visits. It also revealed that although resorts share data with an industry group, they are not readily providing specific injury numbers. Read the story here.
The bet
Scott predicted the ski safety bill, SB 184, would fail 4-1 in the Senate Agriculture and Natural Resources Committee on April 15. Michael predicted it would at least clear the committee.
At stake: bragging rights.
The methodologies
Michael based his prediction on a bit of gumshoe reporting and a vague sense of knowing what might happen next that he calls “his gut.” He interviewed supporters and opponents ahead of the committee hearing and concluded that people discussing the impact of crashes on families — including some tragic deaths — might be persuasive to lawmakers, convincing a small majority to advance the measure to their peers in the Senate for consideration.
Scott made his prediction more scientifically, using a power map created by the National Institute on Money in Politics. The tool uses a database of every lawmakers’ campaign contributions to predict how likely they are to support or oppose a bill. The more money lawmakers get from groups lobbying against a specific bill, the more likely they are to vote ‘no’ on the bill, and vice versa.
Greg Schneider helped design the power mapping tool. He said building the database more than a decade ago was a massive undertaking.
“We still are in a situation where a lot of the states do paper reporting of some sort, but I think we hit the milestone last year where every state at least scans a PDF and puts it online somewhere,” he said.
So Scott looked up public lobbying records for SB 184 and entered that information into the power mapping tool, telling it which groups supported or opposed the ski safety measure.
With this data, the power map predicted every lawmaker in Colorado, except for the bill sponsors, would be likely to vote against it.
That’s because it found that lawmakers had collectively received $5.7 million in campaign contributions from groups actively lobbying against the bill, including the state’s ski industry and a group representing realtors. Supporters of the bill, which included the medical community, only donated $307,000 collectively to lawmakers.
To be clear, donors don’t give money to lawmakers for specific bills. The money tracked in power mapping is an aggregate amount given to all lawmakers from companies and organizations, ranging from Vail Resorts (opposed) to Children’s Hospital Colorado (supporting).
The first test
Lawmakers debated the ski safety bill for about three hours. Forty people signed up to testify.
Don Fisher, a medical doctor and avid skier, said the bill would make Colorado a leader in safety and bring the ski industry in line with the kind of public health data practices that are commonplace in other industries.
“What gets measured, gets managed,” he said.
But Pat Campbell, of Vail Resorts, said the bill would create an "unnecessary burden" on resorts and “would trigger a massive administrative effort to comply,” requiring workers to be directed away from other safety work.
Bill co-sponsor Jessie Danielson said reporters and residents should not have to file open records requests, which can be costly, to learn more about fatalities at ski areas.
“As a skier, I can say if there's a simple, straightforward way to collect the data and require the ski resorts to produce safety plans that's public and transparent, we should do so,” she said. “It may help identify problem areas.”
Before we get to the verdict, what's your prediction?
The verdict
The Senate committee killed the bill, with four senators voting against it and only one — Danielson — voting yes.
The power map correctly predicted how each member would vote, meaning Scott won the bet and bragging rights.
The aftermath
The outcome left Michael and Scott with questions, including whether campaign contributions influenced how the senators voted on Senate Bill 184.
When asked about it, Sen. Kerry Donovan, the committee chair, and Sen. Jerry Sonnenberg, a member of the committee, said there was no connection.
“Absolutely not,” Sonnenberg, a Sterling Republican, said of his ‘no’ vote. “And I would argue that it probably doesn't influence the vote of any legislators.”
Donovan, a Vail Democrat who received $1,000 in campaign contributions from ski interests over the years of running for office, said the monetary support also did not influence her vote “in the slightest” on the bill or on any other legislation.
“You know, I remember that Vail Resorts gave me a check because I was really proud to have the support of my hometown, but I don't know beyond that, if I could tell you who I have and haven't received support of my campaign from,” Donovan said.
She added, “I don’t think anyone can deny the ski industry is a huge driver of our Colorado economy.”
Sen. Jessie Danielson, the bill co-sponsor, said the bill failed because the ski industry lobbied so fiercely against it.
“They refused to negotiate (with us) or even describe why the industry couldn't accept a bill that had to do with transparency around injuries and deaths on their slopes,” she said. “They refused to engage. They refused to work with us on any changes that they would see fit or even any complaints that they had. It was if it was an absolute refusal to even address the issues within the bill. They simply opposed it.”
Danielson claims the ski industry “hired 22 different lobbyists to come to the Capitol and tell us that.”
“And they were successful in defeating the measure in a pretty rapid fashion,” she said.
Danielson was hesitant to react to the power mapping tool accurately predicting how her colleagues would vote on the bill.
“I don't feel comfortable speaking to that,” she said. “I think that, you know, you can draw some conclusions for yourselves.”
Schneider, who helped invent the power mapping tool, accurately predicted how lawmakers would react to questions about the monetary influence, or lack of it, on their votes.
“Most will bristle pretty heavily at the concept that money influences their decision making for lots of reasons, both the practical and the ideological,” he said. “They need to in order to get elected.”
The track record
There is no scientific study on how often the power mapping tool correctly predicts the outcome of votes.
And Schneider admits it has its limits.
So Michael and Scott examined how well it worked for four other major bills in Colorado.
Using the campaign contribution database, the tool accurately predicted that lawmakers would advance an extreme risk protection gun control bill in 2019, a bill allowing governments to name only one finalist in job searches this session, and another on energy efficiency in buildings.
But it failed to predict that lawmakers would also pass a bill in 2019 giving local governments more power to regulate the oil and gas industry.
Schneider cited studies in California and at Princeton University that show campaign contributions and lobbying activity are good predictors of how lawmakers will vote on certain issues. But he stressed that just like on the Colorado bills, there are other variables at play when it comes time to vote.
For example, his tool predicted Colorado Senate President Leroy Garcia would support the extreme risk protection order bill, which allows a judge to temporarily remove weapons from people deemed a threat.
But Garcia was the only Democrat to vote against it in 2019. What the tool did not know is that Garcia’s Democratic predecessor in the Senate was successfully recalled for supporting a ban on high-capacity gun magazines in the wake of the 2012 Aurora theater shooting.
That’s a factor that some old gumshoe reporting could find.
