The odds of a brokered convention are greater than most think.
Six Significant Changes
The superdelegates do not get to vote in the first round this year unless a candidate has a majority. Unlike 2016 when they all went to Hillary, this year they don’t vote until round 2 unless it is already decided.
California is now part of Super Tuesday. In 2016, the California primary was held on June 7. This year, the survivor bias bandwagon effect will be significantly reduced and possibly eliminated.
Following NH there will be two debates, and likely 4 candidates minimum at each. Currently there are six.
This will likely not be a two-way races headed into Super-Tuesday. Elizabeth Warren may have little overall chance, but she does have a chance of getting 15% in many states.
Progressive Split: Bernie Sanders are battling each other for the Progressives. Bernie will get most of this vote, but Warren will likely have have enough money to stay in until the end if she wants.
Bloomberg and Steyer may target a couple of states hard: Texas, Colorado perhaps? They may each pull 15% in a couple of them.
One of the rule changes since 2016 is that superdelegates (uncommitted) cannot vote in the first round unless there is already a clear majority.Specifically
If a single candidate wins at least 2,268 pledged delegates: Superdelegates will be allowed to vote at first ballot, as their influence can not overturn the majority of pledged delegates.
If a single candidate wins 1,886–2,267 pledged delegates: Superdelegates will be barred from voting at first ballot, which solely will be decided by the will of pledged delegates.
If no candidate wins more than 1,885 pledged delegates: This will result in a contested convention, where superdelegates are barred from voting at the first formal ballot, but regain their right to vote for their preferred presidential nominee for all subsequent ballots needed until the delegates reach a majority.
In 2016 Sanders supporters howled, and correctly so, about superdelegate bias for Hillary.
This change alone increases the chances of a brokered convention.
2: Super Tuesday Changes
Compared to 2016, California adds 494 delegates. North Carolina adds 122 delegates. Maine adds 32. And Georgia subtracts 120.
That is a net new 528 delegates that will have at least some survivor bandwagon bias removed.
By survivor bias, I mean increasing the tendency of people to vote for winners as the campaign progresses.
As noted above, the California primary in 2016 was held on June 7. This is a very significant change.
3: Debate Schedule
Six are qualified for the next set of debates.
Billionaire Tom Steyer just qualified. He joins Joe Biden, Bernie Sanders, Elizabeth Warren, and Amy Klobuchar.
There will be two Democratic debates following the New Hampshire primary. The rules will change then, but we do not yet know how.
Those 4-way to 6-way debates can easily take momentum from whoever wins Iowa and New Hampshire.
4: Not a 2-Way Race
The DNC was hoping to narrow the field to two or three. Nope. This will not be a 2016 repeat of Hillary vs Bernie.
Warren is likely to get 15% in many states and might even win Massachusetts.
5: Progressive Split
Bernie is a very strong favorite to beat Warren in the battle for the progressives.
But she has a dedicated following and might easily take 15% of the vote in many states.
Also Warren is good at fundraising. She will likely last to the end.
Please note Tom Steyer's Surprise Surge in SC and NV.
Tom Steyer and Michael Bloomberg are both billionaires. They can self-finance to the end if so desired.
Steyer is highly unlikely to win any states let alone the nomination, but he could easily disrupt South Carolina enough so there is not much momentum for any candidate headed into Super Tuesday.
Bloomberg could conceivably win New York and do very well in Texas on Super Tuesday. By very well I mean 15% or better.
Importance of 15% Threshold
Note that 15% keeps popping up in my analysis.
I discussed why in What Are the Odds of No Winner in the Democratic Primaries?
Republicans have winner-take-all rules in some states but the Democrats generally have some sort of proportional allocation, typically with a 15% threshold.
In California, 35% of the votes are statewide, the rest by district.
According to the California Democratic Party 2020 Delegate Selection Overview, the California Delegation will send a total of 495 delegates the Democratic National Convention comprised of 416 Pledged Delegates and 79 Unpledged Delegates.
Of the pledged delegates, 90 are at-large (statewide) delegates. Another 54 are pledged party leaders (mayors, legislators, state officials, etc.) committed to candidates who get at least 15% of the vote.
(90+54)/416 = 35%
Of the pledged delegates, 326 are allocated by district. The 15% rule comes into play, but at the district level.
Let's run the above math based on a hypothetical California poll that has Bernie at 24%, Warren at 21%, and Biden at 20%.
I am told the poll is real, but there is no reporting on it that I can find.
California Statewide Math
24 + 21 + 20 = 65
Sanders would get 24/65ths of the statewide delegates (37%)
Warren would get 21/65ths of the statewide delegates (32%)
Biden would get 20/65ths of the statewide delegates (31%)
Sanders would get 33 statewide delegates.
Warren would get 29 statewide delegates.
Biden would get 28 statewide delegates.
Those are the statewide allocations.
Pledged Delegate District Math
Sanders would get 24% minimum of 326 district delegates.
Warren would get 21% minimum of 326 district delegates.
Biden would get 20% minimum of 326 district delegates.
Those are approximations for two reasons.
Although there is a 15% minimum, that also applies at the district level. It's possible that Sanders, Warren, or Biden would not get 15% in every district.
It is also possible some candidates get 15% in some districts without hitting the 15% statewide threshold.
Consider the possibility Buttigieg got 15% in half the districts but only 11% statewide. In that case he would get about 7.5% of the 326 or 24 delegates.
Someone is going to "win" but that someone may only have 35-45% of the delegates.
Would that be enough to cause winner bias to kick in?
I don't know. Nor does anyone else. But I doubt it.
Project similar three-way splits across Texas, Illinois, Florida, and New York.
You have no overall winner and thus a "Brokered Convention" even if Bloomberg does not win New York or Warren win Massachusetts.
A brokered convention may seem unlikely, and probably is, but unlikely does not mean zero.
Should it come to a brokered convention, Warren's delegates would likely go to Bernie.
And expect massive howls if Bernie were to have more delegates in round one but lost to Biden when the superdelegates kicked in.
Some say it is too early to be discussing such things, but I would rather discuss this now and throw it away than be in a mad rush to figure everything out at the last minute on Super Tuesday.
Low odds does not mean no odds. I suspect the odds right now are at least 15% and perhaps way higher.